From cfbc4ea907b9eb3af76c7c141659305d971f1ac5 Mon Sep 17 00:00:00 2001 From: Dachuan Shi Date: Mon, 30 Dec 2024 02:04:34 -0500 Subject: [PATCH] Update README.md --- README.md | 40 ++++++++++++++++++---------------------- 1 file changed, 18 insertions(+), 22 deletions(-) diff --git a/README.md b/README.md index d4fc8c0..dceb096 100644 --- a/README.md +++ b/README.md @@ -12,7 +12,7 @@ Webiste
- Visual Reasoning | Image Captioning | Visual QA | Image-Text Retrieval | Image Classification | Image Segmentation + Visual Reasoning | Image Captioning | Visual QA | Image-Text Retrieval | Image Classification | Image Segmentation

@@ -41,23 +41,19 @@ [Text-Image Retrieval](https://github.com/sdc17/UPop#-image-text-and-text-image-retrieval-on-the-flickr30k-dataset-with-clip) • [Flickr30K](https://shannon.cs.illinois.edu/DenotationGraph/) • [CLIP](https://github.com/openai/CLIP) • R@1 | $86.6\rightarrow 80.5_{\color{red}\downarrow 6.1}$ | $856.0\rightarrow 474.3_{\color{ForestGreen}\downarrow 45\\%}$ | $395.7 \rightarrow 201.1_{\color{ForestGreen}\downarrow 49\\%}$ [Classification](https://github.com/sdc17/UPop#-image-classification-on-the-imagenet-dataset) • [ImageNet](https://www.image-net.org/) • [DeiT](https://github.com/facebookresearch/deit) • Acc@1 | $79.9\rightarrow 80.2_{\color{ForestGreen}\uparrow 0.3}$ | $22.0 \rightarrow 15.7_{\color{ForestGreen}\downarrow 29\\%}$ | $4.6 \rightarrow 3.2_{\color{ForestGreen}\downarrow 30\\%}$ [Classification](https://github.com/sdc17/UPop#-image-classification-on-the-imagenet-dataset) • [ImageNet](https://www.image-net.org/) • [DeiT](https://github.com/facebookresearch/deit) • Acc@5 | $95.0 \rightarrow 95.1_{\color{ForestGreen}\uparrow 0.1}$ | $22.0 \rightarrow 15.7_{\color{ForestGreen}\downarrow 29\\%}$ | $4.6 \rightarrow 3.2_{\color{ForestGreen}\downarrow 30\\%}$ - [Segmentation](https://github.com/sdc17/UPop#-image-classification-on-the-imagenet-dataset) • [ADE20K](https://groups.csail.mit.edu/vision/datasets/ADE20K/) • [Segmenter](https://github.com/rstrudel/segmenter) • $\text{mIoU}^s$ | $45.3\rightarrow 45.3_{\color{ForestGreen}\uparrow 0.0}$ | $26.4 \rightarrow 21.5_{\color{ForestGreen}\downarrow 19\\%}$ | $38.6 \rightarrow 30.4_{\color{ForestGreen}\downarrow 21\\%}$ - [Segmentation](https://github.com/sdc17/UPop#-image-classification-on-the-imagenet-dataset) • [ADE20K](https://groups.csail.mit.edu/vision/datasets/ADE20K) • [Segmenter](https://github.com/rstrudel/segmenter) • $\text{mIoU}^m$ | $46.9 \rightarrow 47.1_{\color{ForestGreen}\uparrow 0.2}$ | $26.4 \rightarrow 21.5_{\color{ForestGreen}\downarrow 19\\%}$ | $38.6 \rightarrow 30.4_{\color{ForestGreen}\downarrow 21\\%}$ + [Segmentation](https://github.com/sdc17/UPop#-image-segmentation-on-the-ade20k-dataset) • [ADE20K](https://groups.csail.mit.edu/vision/datasets/ADE20K/) • [Segmenter](https://github.com/rstrudel/segmenter) • $\text{mIoU}^s$ | $45.3\rightarrow 45.3_{\color{ForestGreen}\uparrow 0.0}$ | $26.4 \rightarrow 21.5_{\color{ForestGreen}\downarrow 19\\%}$ | $38.6 \rightarrow 30.4_{\color{ForestGreen}\downarrow 21\\%}$ + [Segmentation](https://github.com/sdc17/UPop#-image-segmentation-on-the-ade20k-dataset) • [ADE20K](https://groups.csail.mit.edu/vision/datasets/ADE20K) • [Segmenter](https://github.com/rstrudel/segmenter) • $\text{mIoU}^m$ | $46.9 \rightarrow 47.1_{\color{ForestGreen}\uparrow 0.2}$ | $26.4 \rightarrow 21.5_{\color{ForestGreen}\downarrow 19\\%}$ | $38.6 \rightarrow 30.4_{\color{ForestGreen}\downarrow 21\\%}$ ### 🥳 What's New -* (Jun 2023), we worked on a new project CrossGET: Cross-Guided Ensemble of Tokens for Accelerating Vision-Language Transformers, which reduces computational costs effectively for accelerating. [[Paper]](https://arxiv.org/pdf/2305.17455.pdf) [[Code]](https://github.com/sdc17/CrossGET) 💡 - -* (Jun 2023), we released the ```implementation```, ```scripts```, ```checkpoints```, and ```logs```. [[Code]](https://github.com/sdc17/UPop) [[Website]](https://dachuanshi.com/UPop-Project/) 🚩 - -* (Apr 2023), our work UPop: Unified and Progressive Pruning for Compressing Vision-Language Transformers was accepted by ICML 2023. [[Paper]](https://proceedings.mlr.press/v202/shi23e/shi23e.pdf) [[ArXiv]](https://arxiv.org/abs/2301.13741) 🎉 +* (Jun 2023), we worked on a new project CrossGET: Cross-Guided Ensemble of Tokens for Accelerating Vision-Language Transformers, which reduces computational costs effectively for accelerating. [[Paper]](https://arxiv.org/pdf/2305.17455.pdf) [[Code]](https://github.com/sdc17/CrossGET) +* (Apr 2023), our work UPop: Unified and Progressive Pruning for Compressing Vision-Language Transformers was accepted by ICML 2023. ### 🏃 Installation The code is tested on `Pytorch==1.11.0`, `cuda==11.3.1`, and `python==3.8.13`. The dependencies can be installed by: ``` conda env create -f environment.yml ``` -The status of installing dependencies: [![build](https://github.com/sdc17/UPop/actions/workflows/build.yml/badge.svg)](https://github.com/sdc17/UPop/actions/workflows/build.yml)