LAVT: Language-Aware Vision Transformer for Referring Image Segmentation

Overview

LAVT: Language-Aware Vision Transformer for Referring Image Segmentation

Where we are ?

12.27 目前和原论文仍有1%左右得差距,但已经力压很多SOTA了

ckpt__448_epoch_25.pth mIoU Overall IoU [email protected]
Refcoco val 70.743 71.671 82.26
Refcoco testA 73.679 74.772 -
Refcoco testB 67.582 67.339 -

12.29 45epoch的结果又上升了大约1%

ckpt__448_epoch_45.pth mIoU Overall IoU
Refcoco val 71.949 72.246
Refcoco testA 74.533 75.467
Refcoco testB 67.849 68.123

the pretrain model will be released soon

对原论文的复现

论文链接: https://arxiv.org/abs/2112.02244

官方实现: https://github.com/yz93/LAVT-RIS

Architecture

Features

  • 将不同模态feature的fusion提前到Image Encoder阶段

  • 思路上对这两篇论文有很多借鉴

    • Vision-Language Transformer and Query Generation for Referring Segmentation

    • Locate then Segment: A Strong Pipeline for Referring Image Segmentation

  • 采用了比较新的主干网络 Swin-Transformer

Usage

详细参数设置可以见args.py

for training

CUDA_VISIBLE_DEVICES=0,1,2,3 python -m torch.distributed.launch --nproc_per_node 4 --master_port 12345 main.py --batch_size 2 --cfg_file configs/swin_base_patch4_window7_224.yaml --size 448

for evaluation

CUDA_VISIBLE_DEVICES=4,5,6,7 python -m torch.distributed.launch --nproc_per_node 4 --master_port 23458 main.py --size 448 --batch_size 1 --resume --eval --type val --eval_mode cat --pretrain ckpt_448_epoch_20.pth --cfg_file configs/swin_base_patch4_window7_224.yaml

*.pth 都放在./checkpoint

for resume from checkpoint

CUDA_VISIBLE_DEVICES=0,1,2,3 python -m torch.distributed.launch --nproc_per_node 4 --master_port 12346 main.py --batch_size 2 --cfg_file configs/swin_base_patch4_window7_224.yaml --size 448 --resume --pretrain ckpt_448_epoch_10.pth

for dataset preparation

please get details from ./data/readme.md

Need to be finished

由于我在复现的时候,官方的code还没有出来,所以一些细节上的设置可能和官方code不同

  • Swin Transformer 我选择的是 swin_base_patch4_window12_384_22k.pth,具体代码可以参考官方代码 https://github.com/microsoft/Swin-Transformer/blob/main/get_started.md 原论文中的图像resize的尺寸是480*480,可是我目前基于官方的代码若想调到这个尺寸,总是会报错,查了一下觉得可能用object detection 的swin transformer的code比较好

    12.27 这个问题目前也已经得到了较好的解决,目前训练用的是 swin_base_patch4_window7_224_22k.pth, 输入图片的尺寸调整到448*448

    解决方案可以参考:

    https://github.com/microsoft/Swin-Transformer/issues/155

  • 原论文中使用的lr_scheduler是polynomial learning rate decay, 没有给出具体的参数手动设置了一下

    12.21 目前来看感觉自己设置的不是很好

    12.27 调整了一下设置,初始学习率的设置真的很重要,特别是根据batch_size 去scale你的 inital learning rate

  • 原论文中的batch_size=32,基于自己的实验我猜想应该是用了8块GPU,每一块的batch_size=4, 由于我第一次写DDP code,训练时发现,程序总是会在RANK0上给其余RANK开辟类似共享显存的东西,导致我无法做到原论文相同的配置,需要改进

  • 仔细观察Refcoco的数据集,会发现一个target会对应好几个sentence,training时我设计的是随机选一个句子,evaluate时感觉应该要把所有句子用上会更好,关于这一点我想了两种evaluate的方法

    目前eval 只能支持 batch_size=1

    • 将所有句子concatenate成为一个句子,送入BERT,Input 形式上就是(Image,cat(sent_1,sent_2,sent_3)) => model => pred

    实验发现这种eval_mode 下的mean IOU 会好不少, overall_IOU 也会好一点

    • 对同一张图片处理多次处理,然后将结果进行平均,Input 形式上就是 ((Image,sent_1),(Image,sent_2),(Image,sent_3)) => model => average(pred_1,pred_2,pred_3)

Visualization

详细见inference.ipynb

input sentences

  1. right girl
  2. closest girl on right

results

Failure cases study

AnalysisFailure.ipynb 提供了一个研究model不work的途径,主要是筛选了IoU < 0.5的case,并在这些case中着重查看了一下IoU < 0.10.4 < IoU < 0.5 的例子

目前我只看了一些有限的failure cases,做了如下总结

  • 模型对于similar,dense object在language guide下定位不精确
  • 模型对于language的理解不分主次
  • refcoco本身标记的一些问题
Owner
zichengsaber
CVer
zichengsaber
ICNet and PSPNet-50 in Tensorflow for real-time semantic segmentation

Real-Time Semantic Segmentation in TensorFlow Perform pixel-wise semantic segmentation on high-resolution images in real-time with Image Cascade Netwo

Oles Andrienko 219 Nov 21, 2022
An implementation for the loss function proposed in Decoupled Contrastive Loss paper.

Decoupled-Contrastive-Learning This repository is an implementation for the loss function proposed in Decoupled Contrastive Loss paper. Requirements P

Ramin Nakhli 71 Dec 04, 2022
Pytorch-Swin-Unet-V2 - a modified version of Swin Unet based on Swin Transfomer V2

Swin Unet V2 Swin Unet V2 is a modified version of Swin Unet arxiv based on Swin

Chenxu Peng 26 Dec 03, 2022
Replication of Pix2Seq with Pretrained Model

Pretrained-Pix2Seq We provide the pre-trained model of Pix2Seq. This version contains new data augmentation. The model is trained for 300 epochs and c

peng gao 51 Nov 22, 2022
Fast, Attemptable Route Planner for Navigation in Known and Unknown Environments

FAR Planner uses a dynamically updated visibility graph for fast replanning. The planner models the environment with polygons and builds a global visi

Fan Yang 346 Dec 30, 2022
Rethinking Nearest Neighbors for Visual Classification

Rethinking Nearest Neighbors for Visual Classification arXiv Environment settings Check out scripts/env_setup.sh Setup data Download the following fin

Menglin Jia 29 Oct 11, 2022
Neural Contours: Learning to Draw Lines from 3D Shapes (CVPR2020)

Neural Contours: Learning to Draw Lines from 3D Shapes This repository contains the PyTorch implementation for CVPR 2020 Paper "Neural Contours: Learn

93 Dec 16, 2022
Official PyTorch implementation of UACANet: Uncertainty Aware Context Attention for Polyp Segmentation

UACANet: Uncertainty Aware Context Attention for Polyp Segmentation Official pytorch implementation of UACANet: Uncertainty Aware Context Attention fo

Taehun Kim 85 Dec 14, 2022
Repositorio de los Laboratorios de Análisis Numérico / Análisis Numérico I de FAMAF, UNC.

Repositorio de los Laboratorios de Análisis Numérico / Análisis Numérico I de FAMAF, UNC. Para los Laboratorios de la materia, vamos a utilizar el len

Luis Biedma 18 Dec 12, 2022
Structural Constraints on Information Content in Human Brain States

Structural Constraints on Information Content in Human Brain States Code accompanying the paper "The information content of brain states is explained

Leon Weninger 3 Sep 07, 2022
An Unpaired Sketch-to-Photo Translation Model

Unpaired-Sketch-to-Photo-Translation We have released our code at https://github.com/rt219/Unsupervised-Sketch-to-Photo-Synthesis This project is the

38 Oct 28, 2022
Repository For Programmers Seeking a platform to show their skills

Programming-Nerds Repository For Programmers Seeking Pull Requests In hacktoberfest ❓ What's Hacktoberfest 2021? Hacktoberfest is the easiest way to g

42 Oct 29, 2022
The source code for 'Noisy-Labeled NER with Confidence Estimation' accepted by NAACL 2021

Kun Liu*, Yao Fu*, Chuanqi Tan, Mosha Chen, Ningyu Zhang, Songfang Huang, Sheng Gao. Noisy-Labeled NER with Confidence Estimation. NAACL 2021. [arxiv]

30 Nov 12, 2022
Official repository of "Investigating Tradeoffs in Real-World Video Super-Resolution"

RealBasicVSR [Paper] This is the official repository of "Investigating Tradeoffs in Real-World Video Super-Resolution, arXiv". This repository contain

Kelvin C.K. Chan 566 Dec 28, 2022
Machine Learning with JAX Tutorials

The purpose of this repo is to make it easy to get started with JAX. It contains my "Machine Learning with JAX" series of tutorials (YouTube videos and Jupyter Notebooks) as well as the content I fou

Aleksa Gordić 372 Dec 28, 2022
Co-mining: Self-Supervised Learning for Sparsely Annotated Object Detection, AAAI 2021.

Co-mining: Self-Supervised Learning for Sparsely Annotated Object Detection This repository is an official implementation of the AAAI 2021 paper Co-mi

MEGVII Research 20 Dec 07, 2022
A lossless neural compression framework built on top of JAX.

Kompressor Branch CI Coverage main (active) main development A neural compression framework built on top of JAX. Install setup.py assumes a compatible

Rosalind Franklin Institute 2 Mar 14, 2022
PyTorch Lightning + Hydra. A feature-rich template for rapid, scalable and reproducible ML experimentation with best practices. ⚡🔥⚡

Lightning-Hydra-Template A clean and scalable template to kickstart your deep learning project 🚀 ⚡ 🔥 Click on Use this template to initialize new re

Łukasz Zalewski 2.1k Jan 09, 2023
High frequency AI based algorithmic trading module.

Flow Flow is a high frequency algorithmic trading module that uses machine learning to self regulate and self optimize for maximum return. The current

59 Dec 14, 2022