AI Rank
EN
← 开源
hfslyc

AdvSemiSeg

Adversarial Learning for Semi-supervised Semantic Segmentation, BMVC 2018

Model DevelopmentArchitecturePython
在 GitHub 打开
增长势头
+024 小时新增 Star0.0%
507
Star
122
Fork
+0
本周
2
贡献者
创建于 2018-02-22 · 更新于 2026-04-24 · 今日第 18157 名
主要开发者
hfslycwasidennis

hfslyc/AdvSemiSeg 排名第 18,157，有 507 个星标。

同类别比较

hfslyc/AdvSemiSeg 排名第 18,157，此处有 8 个同类条目；msracver/Deep-Exemplar-based-Colorization 排名第 18,159。

同类别比较
名称排名指标
shrubb/box-convolutions18,145
语言:
C++
星标数:
508
分叉数:
34
简介:
PyTorch code for the "Deep Neural Networks with Box Convolutions" paper
ynqa/wego18,146
语言:
Go
星标数:
508
分叉数:
45
简介:
Word Embeddings in Go!
AI4Finance-Foundation/FinRL_Podracer18,147
语言:
Python
星标数:
507
分叉数:
121
简介:
Cloud-native Financial Reinforcement Learning
VikParuchuri/textbook_quality18,152
语言:
Python
星标数:
507
分叉数:
45
简介:
Generate textbook-quality synthetic LLM pretraining data
msracver/Deep-Exemplar-based-Colorization18,159
语言:
Python
星标数:
507
分叉数:
75
简介:
The source code of "Deep Exemplar-based Colorization".
ErlerPhilipp/points2surf18,171
语言:
Python
星标数:
506
分叉数:
51
简介:
Learning Implicit Surfaces from Point Clouds (ECCV 2020)
XinJCheng/CSPN18,175
语言:
Python
星标数:
506
分叉数:
90
简介:
Convolutional Spatial Propagation Network
bigscience-workshop/biomedical18,178
语言:
Python
星标数:
506
分叉数:
116
简介:
Tools for curating biomedical training data for large-scale language modeling
关键参数与规格
关键参数与规格
参数值
语言Python
类别Model Development
子类别Architecture
关键词Computer vision
星标数507
分叉数122
贡献者2
创建日期2018-02-22
更新日期2026-04-24
1 天星标数0
7 天星标数0
排名18,157
上次排名17,921
排名变化-236
7 天星标排名16,503
上次 7 天星标排名16,517
7 天星标排名变化14
总星标排名19,156
上次总星标排名19,136
总星标排名变化-20
README 摘录

Adversarial Learning for Semi-supervised Semantic Segmentation This repo is the pytorch implementation of the following paper: Adversarial Learning for Semi-supervised Semantic Segmentation Wei-Chih Hung, Yi-Hsuan Tsai, Yan-Ting Liou, Yen-Yu Lin, and Ming-Hsuan Yang Proceedings of the British Machine Vision Conference (BMVC), 2018. Contact: Wei-Chih Hung (whung8 at ucmerced dot edu) The code are heavily borrowed from a pytorch DeepLab implementation (Link). The baseline model is DeepLabv2-Resnet101 without multiscale training and CRF

post processing, which yields meanIOU 73. Please cite our paper if you find it useful for your research. 3 will cause extra GPU memory on multithread data loader) Installation Clone this repo Place VOC2012 dataset in . For training, you will need the augmented labels (Download). The folder structure should be like: Testing on VOC2012 validation set with pretrained models It will download the pretrained model with 1/8 training data and evaluate on the VOC2012 val set. The colorized images

will be saved in results/ and the detailed class IOU will be saved in results/result. Example visualization results Training on VOC2012 The parameters correspond to those in Table 5 of the paper. To evaluate trained model, execute the following: Changelog 07/24/2018: Update BMVC results

常见问题

记录了多少星标和分叉？

hfslyc/AdvSemiSeg 有 507 个星标和 122 个分叉。

仓库排名是多少？

hfslyc/AdvSemiSeg 排名第 18,157。

记录的语言和类别是什么？

hfslyc/AdvSemiSeg 有 507 个星标，语言为 Python，类别为 Model Development。

哪个仓库排名最接近？

hfslyc/AdvSemiSeg 与 msracver/Deep-Exemplar-based-Colorization 比较，后者排名第 18,159，有 507 个星标。

README

Adversarial Learning for Semi-supervised Semantic Segmentation

This repo is the pytorch implementation of the following paper:

Adversarial Learning for Semi-supervised Semantic Segmentation

Wei-Chih Hung, Yi-Hsuan Tsai, Yan-Ting Liou, Yen-Yu Lin, and Ming-Hsuan Yang

Proceedings of the British Machine Vision Conference (BMVC), 2018.

Contact: Wei-Chih Hung (whung8 at ucmerced dot edu)

The code are heavily borrowed from a pytorch DeepLab implementation (Link). The baseline model is DeepLabv2-Resnet101 without multiscale training and CRF post processing, which yields meanIOU 73.6% on the VOC2012 validation set.

Please cite our paper if you find it useful for your research.

@inproceedings{Hung_semiseg_2018,
  author = {W.-C. Hung and Y.-H. Tsai and Y.-T. Liou and Y.-Y. Lin and M.-H. Yang},
  booktitle = {Proceedings of the British Machine Vision Conference (BMVC)},
  title = {Adversarial Learning for Semi-supervised Semantic Segmentation},
  year = {2018}
}

Prerequisite

  • CUDA/CUDNN
  • pytorch >= 0.2 (We only support 0.4 for evaluation. Will migrate the code to 0.4 soon.)
  • python-opencv >=3.4.0 (3.3 will cause extra GPU memory on multithread data loader)

Installation

  • Clone this repo
git clone https://github.com/hfslyc/AdvSemiSeg.git
  • Place VOC2012 dataset in AdvSemiSeg/dataset/VOC2012. For training, you will need the augmented labels (Download). The folder structure should be like:
AdvSemiSeg/dataset/VOC2012/JPEGImages
                          /SegmentationClassAug

Testing on VOC2012 validation set with pretrained models

python evaluate_voc.py --pretrained-model semi0.125 --save-dir results

It will download the pretrained model with 1/8 training data and evaluate on the VOC2012 val set. The colorized images will be saved in results/ and the detailed class IOU will be saved in results/result.txt. The mean IOU should be around 68.8%.

  • Available --pretrained-model options: semi0.125, semi0.25, semi0.5 , advFull.

Example visualization results

Training on VOC2012

python train.py --snapshot-dir snapshots \
                --partial-data 0.125 \
                --num-steps 20000 \
                --lambda-adv-pred 0.01 \
                --lambda-semi 0.1 --semi-start 5000 --mask-T 0.2

The parameters correspond to those in Table 5 of the paper.

To evaluate trained model, execute the following:

python evaluate_voc.py --restore-from snapshots/VOC_20000.pth \
                       --save-dir results

Changelog

  • 07/24/2018: Update BMVC results
排名相邻
同类项目