1.Graphcore C2 Card performance for image-based deep learning application: A Report ⬇️
Recently, Graphcore has introduced an IPU Processor for accelerating machine learning applications. The architecture of the processor has been designed to achieve state of the art performance on current machine intelligence models for both training and inference.
In this paper, we report on a benchmark in which we have evaluated the performance of IPU processors on deep neural networks for inference. We focus on deep vision models such as ResNeXt. We report the observed latency, throughput and energy efficiency.
2.Pedestrian Models for Autonomous Driving Part I: low level models, from sensing to tracking ⬇️
Autonomous vehicles (AVs) must share space with human pedestrians, both in on-road cases such as cars at pedestrian crossings and off-road cases such as delivery vehicles navigating through crowds on high-streets. Unlike static and kinematic obstacles, pedestrians are active agents with complex, interactive motions. Planning AV actions in the presence of pedestrians thus requires modelling of their probable future behaviour as well as detection and tracking which enable such modelling. This narrative review article is Part I of a pair which together survey the current technology stack involved in this process, organising recent research into a hierarchical taxonomy ranging from low level image detection to high-level psychology models, from the perspective of an AV designer. This self-contained Part I covers the lower levels of this stack, from sensing, through detection and recognition, up to tracking of pedestrians. Technologies at these levels are found to be mature and available as foundations for use in higher level systems such as behaviour modelling, prediction and interaction control.
3.A Quadruplet Loss for Enforcing Semantically Coherent Embeddings in Multi-output Classification Problems ⬇️
This paper describes one objective function for learning semantically coherent feature embeddings in multi-output classification problems, i.e., when the response variables have dimension higher than one. In particular, we consider the problems of identity retrieval and soft biometrics in visual surveillance environments, which have been attracting growing interests. Inspired by the triplet loss function, we propose a generalization of that concept: a quadruplet loss, that 1) defines a metric that analyzes the number of agreeing labels between pairs of elements; and 2) disregards the notion of anchor, replacing d(A1,A2) < d(A1,B) by d(A,B) < d(C,D) distance constraints, according to such perceived semantic similarity between the elements of each pair. Inherited from the triplet loss formulation, our proposal also privileges small distances between positive pairs, but also explicitly enforces that the distances between negative pairs directly correspond to their similarity in terms of the number of agreeing labels. This typically yields feature embeddings with a strong correspondence between the classes centroids and their semantic descriptions, i.e., where elements that share some of the labels are closer to each other in the destiny space than elements with fully disjoint classes membership. Also, in opposition to its triplet counterpart, the proposed loss is not particularly sensitive to the way learning pairs are mined, being agnostic with regard to demanding criteria for mining learning instances (such as the semi-hard pairs of triplet loss). Our experiments were carried out in four different datasets (BIODI, LFW, Megaface and PETA) and validate our assumptions, showing highly promising results.
4.Dynamic Graph Correlation Learning for Disease Diagnosis with Incomplete Labels ⬇️
Disease diagnosis on chest X-ray images is a challenging multi-label classification task. Previous works generally classify the diseases independently on the input image without considering any correlation among diseases. However, such correlation actually exists, for example, Pleural Effusion is more likely to appear when Pneumothorax is present. In this work, we propose a Disease Diagnosis Graph Convolutional Network (DD-GCN) that presents a novel view of investigating the inter-dependency among different diseases by using a dynamic learnable adjacency matrix in graph structure to improve the diagnosis accuracy. To learn more natural and reliable correlation relationship, we feed each node with the image-level individual feature map corresponding to each type of disease. To our knowledge, our method is the first to build a graph over the feature maps with a dynamic adjacency matrix for correlation learning. To further deal with a practical issue of incomplete labels, DD-GCN also utilizes an adaptive loss and a curriculum learning strategy to train the model on incomplete labels. Experimental results on two popular chest X-ray (CXR) datasets show that our prediction accuracy outperforms state-of-the-arts, and the learned graph adjacency matrix establishes the correlation representations of different diseases, which is consistent with expert experience. In addition, we apply an ablation study to demonstrate the effectiveness of each component in DD-GCN.
5.Zooming Slow-Mo: Fast and Accurate One-Stage Space-Time Video Super-Resolution ⬇️
In this paper, we explore the space-time video super-resolution task, which aims to generate a high-resolution (HR) slow-motion video from a low frame rate (LFR), low-resolution (LR) video. A simple solution is to split it into two sub-tasks: video frame interpolation (VFI) and video super-resolution (VSR). However, temporal interpolation and spatial super-resolution are intra-related in this task. Two-stage methods cannot fully take advantage of the natural property. In addition, state-of-the-art VFI or VSR networks require a large frame-synthesis or reconstruction module for predicting high-quality video frames, which makes the two-stage methods have large model sizes and thus be time-consuming. To overcome the problems, we propose a one-stage space-time video super-resolution framework, which directly synthesizes an HR slow-motion video from an LFR, LR video. Rather than synthesizing missing LR video frames as VFI networks do, we firstly temporally interpolate LR frame features in missing LR video frames capturing local temporal contexts by the proposed feature temporal interpolation network. Then, we propose a deformable ConvLSTM to align and aggregate temporal information simultaneously for better leveraging global temporal contexts. Finally, a deep reconstruction network is adopted to predict HR slow-motion video frames. Extensive experiments on benchmark datasets demonstrate that the proposed method not only achieves better quantitative and qualitative performance but also is more than three times faster than recent two-stage state-of-the-art methods, e.g., DAIN+EDVR and DAIN+RBPN.
6.ARMA Nets: Expanding Receptive Field for Dense Prediction ⬇️
Global information is essential for dense prediction problems, whose goal is to compute a discrete or continuous label for each pixel in the images. Traditional convolutional layers in neural networks, originally designed for image classification, are restrictive in these problems since their receptive fields are limited by the filter size. In this work, we propose autoregressive moving-average (ARMA) layer, a novel module in neural networks to allow explicit dependencies of output neurons, which significantly expands the receptive field with minimal extra parameters. We show experimentally that the effective receptive field of neural networks with ARMA layers expands as autoregressive coefficients become larger. In addition, we demonstrate that neural networks with ARMA layers substantially improve the performance of challenging pixel-level video prediction tasks as our model enlarges the effective receptive field.
7.Automatically Searching for U-Net Image Translator Architecture ⬇️
Image translators have been successfully applied to many important low level image processing tasks. However, classical network architecture of image translator like U-Net, is borrowed from other vision tasks like biomedical image segmentation. This straightforward adaptation may not be optimal and could cause redundancy in the network structure. In this paper, we propose an automatic architecture searching method for image translator. By utilizing evolutionary algorithm, we investigate a more efficient network architecture which costs less computation resources and achieves better performance than the original one. Extensive qualitative and quantitative experiments are conducted to demonstrate the effectiveness of the proposed method. Moreover, we transplant the searched network architecture to other datasets which are not involved in the architecture searching procedure. Efficiency of the searched architecture on these datasets further demonstrates the generalization of the method.
8.Object Relational Graph with Teacher-Recommended Learning for Video Captioning ⬇️
Taking full advantage of the information from both vision and language is critical for the video captioning task. Existing models lack adequate visual representation due to the neglect of interaction between object, and sufficient training for content-related words due to long-tailed problems. In this paper, we propose a complete video captioning system including both a novel model and an effective training strategy. Specifically, we propose an object relational graph (ORG) based encoder, which captures more detailed interaction features to enrich visual representation. Meanwhile, we design a teacher-recommended learning (TRL) method to make full use of the successful external language model (ELM) to integrate the abundant linguistic knowledge into the caption model. The ELM generates more semantically similar word proposals which extend the ground-truth words used for training to deal with the long-tailed problem. Experimental evaluations on three benchmarks: MSVD, MSR-VTT and VATEX show the proposed ORG-TRL system achieves state-of-the-art performance. Extensive ablation studies and visualizations illustrate the effectiveness of our system.
9.CookGAN: Meal Image Synthesis from Ingredients ⬇️
In this work we propose a new computational framework, based on generative deep models, for synthesis of photo-realistic food meal images from textual list of its ingredients. Previous works on synthesis of images from text typically rely on pre-trained text models to extract text features, followed by generative neural networks (GAN) aimed to generate realistic images conditioned on the text features. These works mainly focus on generating spatially compact and well-defined categories of objects, such as birds or flowers, but meal images are significantly more complex, consisting of multiple ingredients whose appearance and spatial qualities are further modified by cooking methods. To generate real-like meal images from ingredients, we propose Cook Generative Adversarial Networks (CookGAN), CookGAN first builds an attention-based ingredients-image association model, which is then used to condition a generative neural network tasked with synthesizing meal images. Furthermore, a cycle-consistent constraint is added to further improve image quality and control appearance. Experiments show our model is able to generate meal images corresponding to the ingredients.
10.Learning a Directional Soft Lane Affordance Model for Road Scenes Using Self-Supervision ⬇️
Humans navigate complex environments in an organized yet flexible manner, adapting to the context and implicit social rules. Understanding these naturally learned patterns of behavior is essential for applications such as autonomous vehicles. However, algorithmically defining these implicit rules of human behavior remains difficult. This work proposes a novel self-supervised method for training a probabilistic network model to estimate the regions humans are most likely to drive in as well as a multimodal representation of the inferred direction of travel at each point. The model is trained on individual human trajectories conditioned on a representation of the driving environment. The model is shown to successfully generalize to new road scenes, demonstrating potential for real-world application as a prior for socially acceptable driving behavior in challenging or ambiguous scenarios which are poorly handled by explicit traffic rules.
11.Towards Interpretable Semantic Segmentation via Gradient-weighted Class Activation Mapping ⬇️
Convolutional neural networks have become state-of-the-art in a wide range of image recognition tasks. The interpretation of their predictions, however, is an active area of research. Whereas various interpretation methods have been suggested for image classification, the interpretation of image segmentation still remains largely unexplored. To that end, we propose SEG-GRAD-CAM, a gradient-based method for interpreting semantic segmentation. Our method is an extension of the widely-used Grad-CAM method, applied locally to produce heatmaps showing the relevance of individual pixels for semantic segmentation.
12.Efficient Semantic Video Segmentation with Per-frame Inference ⬇️
In semantic segmentation, most existing real-time deep models trained with each frame independently may produce inconsistent results for a video sequence. Advanced methods take into considerations the correlations in the video sequence, e.g., by propagating the results to the neighboring frames using optical flow, or extracting the frame representations with other frames, which may lead to inaccurate results or unbalanced latency. In this work, we process efficient semantic video segmentation in a per-frame fashion during the inference process. Different from previous per-frame models, we explicitly consider the temporal consistency among frames as extra constraints during the training process and embed the temporal consistency into the segmentation network. Therefore, in the inference process, we can process each frame independently with no latency, and improve the temporal consistency with no extra computational cost and post-processing. We employ compact models for real-time execution. To narrow the performance gap between compact models and large models, new knowledge distillation methods are designed. Our results outperform previous keyframe based methods with a better trade-off between the accuracy and the inference speed on popular benchmarks, including the Cityscapes and Camvid. The temporal consistency is also improved compared with corresponding baselines which are trained with each frame independently. Code is available at: this https URL
13.Deform-GAN:An Unsupervised Learning Model for Deformable Registration ⬇️
Deformable registration is one of the most challenging task in the field of medical image analysis, especially for the alignment between different sequences and modalities. In this paper, a non-rigid registration method is proposed for 3D medical images leveraging unsupervised learning. To the best of our knowledge, this is the first attempt to introduce gradient loss into deep-learning-based registration. The proposed gradient loss is robust across sequences and modals for large deformation. Besides, adversarial learning approach is used to transfer multi-modal similarity to mono-modal similarity and improve the precision. Neither ground-truth nor manual labeling is required during training. We evaluated our network on a 3D brain registration task comprehensively. The experiments demonstrate that the proposed method can cope with the data which has non-functional intensity relations, noise and blur. Our approach outperforms other methods especially in accuracy and speed.
14.Performance Evaluation of Deep Generative Models for Generating Hand-Written Character Images ⬇️
There have been many work in the literature on generation of various kinds of images such as Hand-Written characters (MNIST dataset), scene images (CIFAR-10 dataset), various objects images (ImageNet dataset), road signboard images (SVHN dataset) etc. Unfortunately, there have been very limited amount of work done in the domain of document image processing. Automatic image generation can lead to the enormous increase of labeled datasets with the help of only limited amount of labeled data. Various kinds of Deep generative models can be primarily divided into two categories. First category is auto-encoder (AE) and the second one is Generative Adversarial Networks (GANs). In this paper, we have evaluated various kinds of AE as well as GANs and have compared their performances on hand-written digits dataset (MNIST) and also on historical hand-written character dataset of Indonesian BALI language. Moreover, these generated characters are recognized by using character recognition tool for calculating the statistical performance of these generated characters with respect to original character images.
15.Disentangling Image Distortions in Deep Feature Space ⬇️
Previous literature suggests that perceptual similarity is an emergent property shared across deep visual representations. Experiments conducted on a dataset of human-judged image distortions have proven that deep features outperform, by a large margin, classic perceptual metrics. In this work we take a further step in the direction of a broader understanding of such property by analyzing the capability of deep visual representations to intrinsically characterize different types of image distortions. To this end, we firstly generate a number of synthetically distorted images by applying three mainstream distortion types to the LIVE database and then we analyze the features extracted by different layers of different Deep Network architectures. We observe that a dimension-reduced representation of the features extracted from a given layer permits to efficiently separate types of distortions in the feature space. Moreover, each network layer exhibits a different ability to separate between different types of distortions, and this ability varies according to the network architecture. As a further analysis, we evaluate the exploitation of features taken from the layer that better separates image distortions for: i) reduced-reference image quality assessment, and ii) distortion types and severity levels characterization on both single and multiple distortion databases. Results achieved on both tasks suggest that deep visual representations can be unsupervisedly employed to efficiently characterize various image distortions.
16.Force-Ultrasound Fusion:Bringing Spine Robotic-US to the Next "Level" ⬇️
Spine injections are commonly performed in several clinical procedures. The localization of the target vertebral level (i.e. the position of a vertebra in a spine) is typically done by back palpation or under X-ray guidance, yielding either higher chances of procedure failure or exposure to ionizing radiation. Preliminary studies have been conducted in the literature, suggesting that ultrasound imaging may be a precise and safe alternative to X-ray for spine level detection. However, ultrasound data are noisy and complicated to interpret. In this study, a robotic-ultrasound approach for automatic vertebral level detection is introduced. The method relies on the fusion of ultrasound and force data, thus providing both "tactile" and visual feedback during the procedure, which results in higher performances in presence of data corruption. A robotic arm automatically scans the volunteer's back along the spine by using force-ultrasound data to locate vertebral levels. The occurrences of vertebral levels are visible on the force trace as peaks, which are enhanced by properly controlling the force applied by the robot on the patient back. Ultrasound data are processed with a Deep Learning method to extract a 1D signal modelling the probabilities of having a vertebra at each location along the spine. Processed force and ultrasound data are fused using a 1D Convolutional Network to compute the location of the vertebral levels. The method is compared to pure image and pure force-based methods for vertebral level counting, showing improved performance. In particular, the fusion method is able to correctly classify 100% of the vertebral levels in the test set, while pure image and pure force-based method could only classify 80% and 90% vertebrae, respectively. The potential of the proposed method is evaluated in an exemplary simulated clinical application.
17.Controllable Descendant Face Synthesis ⬇️
Kinship face synthesis is an interesting topic raised to answer questions like "what will your future children look like?". Published approaches to this topic are limited. Most of the existing methods train models for one-versus-one kin relation, which only consider one parent face and one child face by directly using an auto-encoder without any explicit control over the resemblance of the synthesized face to the parent face. In this paper, we propose a novel method for controllable descendant face synthesis, which models two-versus-one kin relation between two parent faces and one child face. Our model consists of an inheritance module and an attribute enhancement module, where the former is designed for accurate control over the resemblance between the synthesized face and parent faces, and the latter is designed for control over age and gender. As there is no large scale database with father-mother-child kinship annotation, we propose an effective strategy to train the model without using the ground truth descendant faces. No carefully designed image pairs are required for learning except only age and gender labels of training faces. We conduct comprehensive experimental evaluations on three public benchmark databases, which demonstrates encouraging results.
18.Adversarial Attack on Deep Product Quantization Network for Image Retrieval ⬇️
Deep product quantization network (DPQN) has recently received much attention in fast image retrieval tasks due to its efficiency of encoding high-dimensional visual features especially when dealing with large-scale datasets. Recent studies show that deep neural networks (DNNs) are vulnerable to input with small and maliciously designed perturbations (a.k.a., adversarial examples). This phenomenon raises the concern of security issues for DPQN in the testing/deploying stage as well. However, little effort has been devoted to investigating how adversarial examples affect DPQN. To this end, we propose product quantization adversarial generation (PQ-AG), a simple yet effective method to generate adversarial examples for product quantization based retrieval systems. PQ-AG aims to generate imperceptible adversarial perturbations for query images to form adversarial queries, whose nearest neighbors from a targeted product quantizaiton model are not semantically related to those from the original queries. Extensive experiments show that our PQ-AQ successfully creates adversarial examples to mislead targeted product quantization retrieval models. Besides, we found that our PQ-AG significantly degrades retrieval performance in both white-box and black-box settings.
19.PuzzleNet: Scene Text Detection by Segment Context Graph Learning ⬇️
Recently, a series of decomposition-based scene text detection methods has achieved impressive progress by decomposing challenging text regions into pieces and linking them in a bottom-up manner. However, most of them merely focus on linking independent text pieces while the context information is underestimated. In the puzzle game, the solver often put pieces together in a logical way according to the contextual information of each piece, in order to arrive at the correct solution. Inspired by it, we propose a novel decomposition-based method, termed Puzzle Networks (PuzzleNet), to address the challenging scene text detection task in this work. PuzzleNet consists of the Segment Proposal Network (SPN) that predicts the candidate text segments fitting arbitrary shape of text region, and the two-branch Multiple-Similarity Graph Convolutional Network (MSGCN) that models both appearance and geometry correlations between each segment to its contextual ones. By building segments as context graphs, MSGCN effectively employs segment context to predict combinations of segments. Final detections of polygon shape are produced by merging segments according to the predicted combinations. Evaluations on three benchmark datasets, ICDAR15, MSRA-TD500 and SCUT-CTW1500, have demonstrated that our method can achieve better or comparable performance than current state-of-the-arts, which is beneficial from the exploitation of segment context graph.
20.Unsupervised Temporal Video Segmentation as an Auxiliary Task for Predicting the Remaining Surgery Duration ⬇️
Estimating the remaining surgery duration (RSD) during surgical procedures can be useful for OR planning and anesthesia dose estimation. With the recent success of deep learning-based methods in computer vision, several neural network approaches have been proposed for fully automatic RSD prediction based solely on visual data from the endoscopic camera. We investigate whether RSD prediction can be improved using unsupervised temporal video segmentation as an auxiliary learning task. As opposed to previous work, which presented supervised surgical phase recognition as auxiliary task, we avoid the need for manual annotations by proposing a similar but unsupervised learning objective which clusters video sequences into temporally coherent segments. In multiple experimental setups, results obtained by learning the auxiliary task are incorporated into a deep RSD model through feature extraction, pretraining or regularization. Further, we propose a novel loss function for RSD training which attempts to counteract unfavorable characteristics of the RSD ground truth. Using our unsupervised method as an auxiliary task for RSD training, we outperform other self-supervised methods and are comparable to the supervised state-of-the-art. Combined with the novel RSD loss, we slightly outperform the supervised approach.
21.Rethinking the Route Towards Weakly Supervised Object Localization ⬇️
Weakly supervised object localization (WSOL) aims to localize objects with only image-level labels. Previous methods often try to utilize feature maps and classification weights to localize objects using image level annotations indirectly. In this paper, we demonstrate that weakly supervised object localization should be divided into two parts: class-agnostic object localization and object classification. For class-agnostic object localization, we should use class-agnostic methods to generate noisy pseudo annotations and then perform bounding box regression on them without class labels. We propose the pseudo supervised object localization (PSOL) method as a new way to solve WSOL. Our PSOL models have good transferability across different datasets without fine-tuning. With generated pseudo bounding boxes, we achieve 58.00% localization accuracy on ImageNet and 74.74% localization accuracy on CUB-200, which have a large edge over previous models.
22.Refined Gate: A Simple and Effective Gating Mechanism for Recurrent Units ⬇️
Recurrent neural network (RNN) has been widely studied in sequence learning tasks, while the mainstream models (e.g., LSTM and GRU) rely on the gating mechanism (in control of how information flows between hidden states). However, the vanilla gates in RNN (e.g. the input gate in LSTM) suffer from the problem of gate undertraining mainly due to the saturating activation functions, which may result in failures of learning gating roles and thus the weak performance. In this paper, we propose a new gating mechanism within general gated recurrent neural networks to handle this issue. Specifically, the proposed gates directly short connect the extracted input features to the outputs of vanilla gates, denoted as refined gates. The refining mechanism allows enhancing gradient back-propagation as well as extending the gating activation scope, which, although simple, can guide RNN to reach possibly deeper minima. We verify the proposed gating mechanism on three popular types of gated RNNs including LSTM, GRU and MGU. Extensive experiments on 3 synthetic tasks, 3 language modeling tasks and 5 scene text recognition benchmarks demonstrate the effectiveness of our method.
23.Self-supervised Image Enhancement Network: Training with Low Light Images Only ⬇️
This paper proposes a self-supervised low light image enhancement method based on deep learning. Inspired by information entropy theory and Retinex model, we proposed a maximum entropy based Retinex model. With this model, a very simple network can separate the illumination and reflectance, and the network can be trained with low light images only. We introduce a constraint that the maximum channel of the reflectance conforms to the maximum channel of the low light image and its entropy should be largest in our model to achieve self-supervised learning. Our model is very simple and does not rely on any well-designed data set (even one low light image can complete the training). The network only needs minute-level training to achieve image enhancement. It can be proved through experiments that the proposed method has reached the state-of-the-art in terms of processing speed and effect.
24.Generalized ODIN: Detecting Out-of-distribution Image without Learning from Out-of-distribution Data ⬇️
Deep neural networks have attained remarkable performance when applied to data that comes from the same distribution as that of the training set, but can significantly degrade otherwise. Therefore, detecting whether an example is out-of-distribution (OoD) is crucial to enable a system that can reject such samples or alert users. Recent works have made significant progress on OoD benchmarks consisting of small image datasets. However, many recent methods based on neural networks rely on training or tuning with both in-distribution and out-of-distribution data. The latter is generally hard to define a-priori, and its selection can easily bias the learning. We base our work on a popular method ODIN, proposing two strategies for freeing it from the needs of tuning with OoD data, while improving its OoD detection performance. We specifically propose to decompose confidence scoring as well as a modified input pre-processing method. We show that both of these significantly help in detection performance. Our further analysis on a larger scale image dataset shows that the two types of distribution shifts, specifically semantic shift and non-semantic shift, present a significant difference in the difficulty of the problem, providing an analysis of when ODIN-like strategies do or do not work.
25.Adversarial Ranking Attack and Defense ⬇️
Deep Neural Network (DNN) classifiers are vulnerable to adversarial attack, where an imperceptible perturbation could result in misclassification. However, the vulnerability of DNN-based image ranking systems remains under-explored. In this paper, we propose two attacks against deep ranking systems, i.e., Candidate Attack and Query Attack, that can raise or lower the rank of chosen candidates by adversarial perturbations. Specifically, the expected ranking order is first represented as a set of inequalities, and then a triplet-like objective function is designed to obtain the optimal perturbation. Conversely, a defense method is also proposed to improve the ranking system robustness, which can mitigate all the proposed attacks simultaneously. Our adversarial ranking attacks and defense are evaluated on datasets including MNIST, Fashion-MNIST, and Stanford-Online-Products. Experimental results demonstrate that a typical deep ranking system can be effectively compromised by our attacks. Meanwhile, the system robustness can be moderately improved with our defense. Furthermore, the transferable and universal properties of our adversary illustrate the possibility of realistic black-box attack.
26.Generalized Product Quantization Network for Semi-supervised Hashing ⬇️
Learning to hash has achieved great success in image retrieval due to its low storage cost and fast search speed. In recent years, hashing methods that take advantage of deep learning have come into the spotlight with some positive outcomes. However, these approaches do not meet expectations unless expensive label information is sufficient. To resolve this issue, we propose the first quantization-based semi-supervised hashing scheme: Generalized Product Quantization (\textbf{GPQ}) network. We design a novel metric learning strategy that preserves semantic similarity between labeled data, and employ entropy regularization term to fully exploit inherent potentials of unlabeled data. Our solution increases the generalization capacity of the hash function, which allows overcoming previous limitations in the retrieval community. Extensive experimental results demonstrate that GPQ yields state-of-the-art performance on large-scale real image benchmark datasets.
27.Learning Light Field Angular Super-Resolution via a Geometry-Aware Network ⬇️
The acquisition of light field images with high angular resolution is costly. Although many methods have been proposed to improve the angular resolution of a sparsely-sampled light field, they always focus on the light field with a small baseline, which is captured by a consumer light field camera. By making full use of the intrinsic \textit{geometry} information of light fields, in this paper we propose an end-to-end learning-based approach aiming at angularly super-resolving a sparsely-sampled light field with a large baseline. Our model consists of two learnable modules and a physically-based module. Specifically, it includes a depth estimation module for explicitly modeling the scene geometry, a physically-based warping for novel views synthesis, and a light field blending module specifically designed for light field reconstruction. Moreover, we introduce a novel loss function to promote the preservation of the light field parallax structure. Experimental results over various light field datasets including large baseline light field images demonstrate the significant superiority of our method when compared with state-of-the-art ones, i.e., our method improves the PSNR of the second best method up to 2 dB in average, while saves the execution time 48$\times$. In addition, our method preserves the light field parallax structure better.
28.Multi-Attribute Guided Painting Generation ⬇️
Controllable painting generation plays a pivotal role in image stylization. Currently, the control way of style transfer is subject to exemplar-based reference or a random one-hot vector guidance. Few works focus on decoupling the intrinsic properties of painting as control conditions, e.g., artist, genre and period. Under this circumstance, we propose a novel framework adopting multiple attributes from the painting to control the stylized results. An asymmetrical cycle structure is equipped to preserve the fidelity, associating with style preserving and attribute regression loss to keep the unique distinction of colors and textures between domains. Several qualitative and quantitative results demonstrate the effect of the combinations of multiple attributes and achieve satisfactory performance.
29.Back to the Future: Joint Aware Temporal Deep Learning 3D Human Pose Estimation ⬇️
We propose a new deep learning network that introduces a deeper CNN channel filter and constraints as losses to reduce joint position and motion errors for 3D video human body pose estimation. Our model outperforms the previous best result from the literature based on mean per-joint position error, velocity error, and acceleration errors on the Human 3.6M benchmark corresponding to a new state-of-the-art mean error reduction in all protocols and motion metrics. Mean per joint error is reduced by 1%, velocity error by 7% and acceleration by 13% compared to the best results from the literature. Our contribution increasing positional accuracy and motion smoothness in video can be integrated with future end to end networks without increasing network complexity. Our model and code are available at this https URL
Keywords: 3D, human, image, pose, action, detection, object, video, visual, supervised, joint, kinematic
30.Super-Resolving Commercial Satellite Imagery Using Realistic Training Data ⬇️
In machine learning based single image super-resolution, the degradation model is embedded in training data generation. However, most existing satellite image super-resolution methods use a simple down-sampling model with a fixed kernel to create training images. These methods work fine on synthetic data, but do not perform well on real satellite images. We propose a realistic training data generation model for commercial satellite imagery products, which includes not only the imaging process on satellites but also the post-process on the ground. We also propose a convolutional neural network optimized for satellite images. Experiments show that the proposed training data generation model is able to improve super-resolution performance on real satellite images.
31.Transfer Learning from Synthetic to Real-Noise Denoising with Adaptive Instance Normalization ⬇️
Real-noise denoising is a challenging task because the statistics of real-noise do not follow the normal distribution, and they are also spatially and temporally changing. In order to cope with various and complex real-noise, we propose a well-generalized denoising architecture and a transfer learning scheme. Specifically, we adopt an adaptive instance normalization to build a denoiser, which can regularize the feature map and prevent the network from overfitting to the training set. We also introduce a transfer learning scheme that transfers knowledge learned from synthetic-noise data to the real-noise denoiser. From the proposed transfer learning, the synthetic-noise denoiser can learn general features from various synthetic-noise data, and the real-noise denoiser can learn the real-noise characteristics from real data. From the experiments, we find that the proposed denoising method has great generalization ability, such that our network trained with synthetic-noise achieves the best performance for Darmstadt Noise Dataset (DND) among the methods from published papers. We can also see that the proposed transfer learning scheme robustly works for real-noise images through the learning with a very small number of labeled data.
32.Style Transfer for Light Field Photography ⬇️
As light field images continue to increase in use and application, it becomes necessary to adapt existing image processing methods to this unique form of photography. In this paper we explore methods for applying neural style transfer to light field images. Feed-forward style transfer networks provide fast, high-quality results for monocular images, but no such networks exist for full light field images. Because of the size of these images, current light field data sets are small and are insufficient for training purely feed-forward style-transfer networks from scratch. Thus, it is necessary to adapt existing monocular style transfer networks in a way that allows for the stylization of each view of the light field while maintaining visual consistencies between views. Instead, the proposed method backpropagates the loss through the network, and the process is iterated to optimize (essentially overfit) the resulting stylization for a single light field image alone. The network architecture allows for the incorporation of pre-trained fast monocular stylization networks while avoiding the need for a large light field training set.
33.Geometric Fusion via Joint Delay Embeddings ⬇️
We introduce geometric and topological methods to develop a new framework for fusing multi-sensor time series. This framework consists of two steps: (1) a joint delay embedding, which reconstructs a high-dimensional state space in which our sensors correspond to observation functions, and (2) a simple orthogonalization scheme, which accounts for tangencies between such observation functions, and produces a more diversified geometry on the embedding space. We conclude with some synthetic and real-world experiments demonstrating that our framework outperforms traditional metric fusion methods.
34.Unsupervised Semantic Attribute Discovery and Control in Generative Models ⬇️
This work focuses on the ability to control via latent space factors semantic image attributes in generative models, and the faculty to discover mappings from factors to attributes in an unsupervised fashion. The discovery of controllable semantic attributes is of special importance, as it would facilitate higher level tasks such as unsupervised representation learning to improve anomaly detection, or the controlled generation of novel data for domain shift and imbalanced datasets. The ability to control semantic attributes is related to the disentanglement of latent factors, which dictates that latent factors be "uncorrelated" in their effects. Unfortunately, despite past progress, the connection between control and disentanglement remains, at best, confused and entangled, requiring clarifications we hope to provide in this work. To this end, we study the design of algorithms for image generation that allow unsupervised discovery and control of semantic attributes.We make several contributions: a) We bring order to the concepts of control and disentanglement, by providing an analytical derivation that connects mutual information maximization, which promotes attribute control, to total correlation minimization, which relates to disentanglement. b) We propose hybrid generative model architectures that use mutual information maximization with multi-scale style transfer. c) We introduce a novel metric to characterize the performance of semantic attributes control. We report experiments that appear to demonstrate, quantitatively and qualitatively, the ability of the proposed model to perform satisfactory control while still preserving competitive visual quality. We compare to other state of the art methods (e.g., Frechet inception distance (FID)= 9.90 on CelebA and 4.52 on EyePACS).
35.CLARA: Clinical Report Auto-completion ⬇️
Generating clinical reports from raw recordings such as X-rays and electroencephalogram (EEG) is an essential and routine task for doctors. However, it is often time-consuming to write accurate and detailed reports. Most existing methods try to generate the whole reports from the raw input with limited success because 1) generated reports often contain errors that need manual review and correction, 2) it does not save time when doctors want to write additional information into the report, and 3) the generated reports are not customized based on individual doctors' preference. We propose {\it CL}inic{\it A}l {\it R}eport {\it A}uto-completion (CLARA), an interactive method that generates reports in a sentence by sentence fashion based on doctors' anchor words and partially completed sentences. CLARA searches for most relevant sentences from existing reports as the template for the current report. The retrieved sentences are sequentially modified by combining with the input feature representations to create the final report. In our experimental evaluation, CLARA achieved 0.393 CIDEr and 0.248 BLEU-4 on X-ray reports and 0.482 CIDEr and 0.491 BLEU-4 for EEG reports for sentence-level generation, which is up to 35% improvement over the best baseline. Also via our qualitative evaluation, CLARA is shown to produce reports which have a significantly higher level of approval by doctors in a user study (3.74 out of 5 for CLARA vs 2.52 out of 5 for the baseline).
36.Inceptive Event Time-Surfaces for Object Classification Using Neuromorphic Cameras ⬇️
This paper presents a novel fusion of low-level approaches for dimensionality reduction into an effective approach for high-level objects in neuromorphic camera data called Inceptive Event Time-Surfaces (IETS). IETSs overcome several limitations of conventional time-surfaces by increasing robustness to noise, promoting spatial consistency, and improving the temporal localization of (moving) edges. Combining IETS with transfer learning improves state-of-the-art performance on the challenging problem of object classification utilizing event camera data.
37.Revisiting Ensembles in an Adversarial Context: Improving Natural Accuracy ⬇️
A necessary characteristic for the deployment of deep learning models in real world applications is resistance to small adversarial perturbations while maintaining accuracy on non-malicious inputs. While robust training provides models that exhibit better adversarial accuracy than standard models, there is still a significant gap in natural accuracy between robust and non-robust models which we aim to bridge. We consider a number of ensemble methods designed to mitigate this performance difference. Our key insight is that model trained to withstand small attacks, when ensembled, can often withstand significantly larger attacks, and this concept can in turn be leveraged to optimize natural accuracy. We consider two schemes, one that combines predictions from several randomly initialized robust models, and the other that fuses features from robust and standard models.
38.Region of Interest Identification for Brain Tumors in Magnetic Resonance Images ⬇️
Glioma is a common type of brain tumor, and accurate detection of it plays a vital role in the diagnosis and treatment process. Despite advances in medical image analyzing, accurate tumor segmentation in brain magnetic resonance (MR) images remains a challenge due to variations in tumor texture, position, and shape. In this paper, we propose a fast, automated method, with light computational complexity, to find the smallest bounding box around the tumor region. This region-of-interest can be used as a preprocessing step in training networks for subregion tumor segmentation. By adopting the outputs of this algorithm, redundant information is removed; hence the network can focus on learning notable features related to subregions' classes. The proposed method has six main stages, in which the brain segmentation is the most vital step. Expectation-maximization (EM) and K-means algorithms are used for brain segmentation. The proposed method is evaluated on the BraTS 2015 dataset, and the average gained DICE score is 0.73, which is an acceptable result for this application.
39.Unpaired Image Super-Resolution using Pseudo-Supervision ⬇️
In most studies on learning-based image super-resolution (SR), the paired training dataset is created by downscaling high-resolution (HR) images with a predetermined operation (e.g., bicubic). However, these methods fail to super-resolve real-world low-resolution (LR) images, for which the degradation process is much more complicated and unknown. In this paper, we propose an unpaired SR method using a generative adversarial network that does not require a paired/aligned training dataset. Our network consists of an unpaired kernel/noise correction network and a pseudo-paired SR network. The correction network removes noise and adjusts the kernel of the inputted LR image; then, the corrected clean LR image is upscaled by the SR network. In the training phase, the correction network also produces a pseudo-clean LR image from the inputted HR image, and then a mapping from the pseudo-clean LR image to the inputted HR image is learned by the SR network in a paired manner. Because our SR network is independent of the correction network, well-studied existing network architectures and pixel-wise loss functions can be integrated with the proposed framework. Experiments on diverse datasets show that the proposed method is superior to existing solutions to the unpaired SR problem.
40.CheXpedition: Investigating Generalization Challenges for Translation of Chest X-Ray Algorithms to the Clinical Setting ⬇️
Although there have been several recent advances in the application of deep learning algorithms to chest x-ray interpretation, we identify three major challenges for the translation of chest x-ray algorithms to the clinical setting. We examine the performance of the top 10 performing models on the CheXpert challenge leaderboard on three tasks: (1) TB detection, (2) pathology detection on photos of chest x-rays, and (3) pathology detection on data from an external institution. First, we find that the top 10 chest x-ray models on the CheXpert competition achieve an average AUC of 0.851 on the task of detecting TB on two public TB datasets without fine-tuning or including the TB labels in training data. Second, we find that the average performance of the models on photos of x-rays (AUC = 0.916) is similar to their performance on the original chest x-ray images (AUC = 0.924). Third, we find that the models tested on an external dataset either perform comparably to or exceed the average performance of radiologists. We believe that our investigation will inform rapid translation of deep learning algorithms to safe and effective clinical decision support tools that can be validated prospectively with large impact studies and clinical trials.
41.Invariance vs. Robustness of Neural Networks ⬇️
We study the performance of neural network models on random geometric transformations and adversarial perturbations. Invariance means that the model's prediction remains unchanged when a geometric transformation is applied to an input. Adversarial robustness means that the model's prediction remains unchanged after small adversarial perturbations of an input. In this paper, we show a quantitative trade-off between rotation invariance and robustness. We empirically study the following two cases: (a) change in adversarial robustness as we improve only the invariance of equivariant models via training augmentation, (b) change in invariance as we improve only the adversarial robustness using adversarial training. We observe that the rotation invariance of equivariant models (StdCNNs and GCNNs) improves by training augmentation with progressively larger random rotations but while doing so, their adversarial robustness drops progressively, and very significantly on MNIST. We take adversarially trained LeNet and ResNet models which have good
$L_\infty$ adversarial robustness on MNIST and CIFAR-10, respectively, and observe that adversarial training with progressively larger perturbations results in a progressive drop in their rotation invariance profiles. Similar to the trade-off between accuracy and robustness known in previous work, we give a theoretical justification for the invariance vs. robustness trade-off observed in our experiments.
42.From Seeing to Moving: A Survey on Learning for Visual Indoor Navigation (VIN) ⬇️
Visual Indoor Navigation (VIN) task has drawn increasing attentions from the data-driven machine learning communities especially with the recent reported success from learning-based methods. Due to the innate complexity of this task, researchers have tried approaching the problem from a variety of different angles, the full scope of which has not yet been captured within an overarching report. In this survey, we discuss the representative work of learning-based approaches for visual navigation and its related tasks. Firstly, we summarize the current work in terms of task representations and applied methods along with their properties. We then further identify and discuss lingering issues impeding the performance of VIN tasks and motivate future research in these key areas worth exploring in the future for the community.
43.Deep Learning and Statistical Models for Time-Critical Pedestrian Behaviour Prediction ⬇️
The time it takes for a classifier to make an accurate prediction can be crucial in many behaviour recognition problems. For example, an autonomous vehicle should detect hazardous pedestrian behaviour early enough for it to take appropriate measures. In this context, we compare the switching linear dynamical system (SLDS) and a three-layered bi-directional long short-term memory (LSTM) neural network, which are applied to infer pedestrian behaviour from motion tracks. We show that, though the neural network model achieves an accuracy of 80%, it requires long sequences to achieve this (100 samples or more). The SLDS, has a lower accuracy of 74%, but it achieves this result with short sequences (10 samples). To our knowledge, such a comparison on sequence length has not been considered in the literature before. The results provide a key intuition of the suitability of the models in time-critical problems.
44.End-to-End Models for the Analysis of System 1 and System 2 Interactions based on Eye-Tracking Data ⬇️
While theories postulating a dual cognitive system take hold, quantitative confirmations are still needed to understand and identify interactions between the two systems or conflict events. Eye movements are among the most direct markers of the individual attentive load and may serve as an important proxy of information. In this work we propose a computational method, within a modified visual version of the well-known Stroop test, for the identification of different tasks and potential conflicts events between the two systems through the collection and processing of data related to eye movements. A statistical analysis shows that the selected variables can characterize the variation of attentive load within different scenarios. Moreover, we show that Machine Learning techniques allow to distinguish between different tasks with a good classification accuracy and to investigate more in depth the gaze dynamics.