Dark Mode

Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

Sign up
Appearance settings

Easy-to-use image segmentation library with awesome pre-trained model zoo, supporting wide-range of practical tasks in Semantic Segmentation, Interactive Segmentation, Panoptic Segmentation, Image Matting, 3D Segmentation, etc.

License

Notifications You must be signed in to change notification settings

GoldenStain/PaddleSeg

Repository files navigation

Jian Ti Zhong Wen | English

Fei Jiang Gao Xing Neng Tu Xiang Fen Ge Kai Fa Tao Jian ,Duan Dao Duan Wan Cheng Cong Xun Lian Dao Bu Shu De Quan Liu Cheng Tu Xiang Fen Ge Ying Yong .

Zui Xin Dong Tai

  • [2024-11-05] Tian Jia Yu Yi Fen Ge Ling Yu Di Dai Ma Quan Liu Cheng Kai Fa Neng Li :
    • Fei Jiang Di Dai Ma Kai Fa Gong Ju PaddleX,Yi Tuo Yu PaddleSegDe Xian Jin Ji Zhu ,Zhi Chi Liao Tu Xiang Fen Ge Ling Yu De Di Dai Ma Quan Liu Cheng Kai Fa Neng Li :

      • Mo Xing Feng Fu Yi Jian Diao Yong :Jiang Tong Yong Yu Yi Fen Ge He Tu Xiang Yi Chang Jian Ce She Ji De 19Ge Mo Xing Zheng He Wei 2Tiao Mo Xing Chan Xian ,Tong Guo Ji Jian De Python APIYi Jian Diao Yong ,Kuai Su Ti Yan Mo Xing Xiao Guo . Ci Wai ,Tong Yi Tao API,Ye Zhi Chi Tu Xiang Fen Lei , Mu Biao Jian Ce , Wen Ben Tu Xiang Zhi Neng Fen Xi , Tong Yong OCR, Shi Xu Yu Ce Deng Gong Ji 200+Mo Xing ,Xing Cheng 20+Dan Gong Neng Mo Kuai ,Fang Bian Kai Fa Zhe Jin Xing Mo Xing Zu He Shi Yong .
      • Ti Gao Xiao Lu Jiang Di Men Jian :Ti Gong Ji Yu Tong Yi Ming Ling He Tu Xing Jie Mian Liang Chong Fang Shi ,Shi Xian Mo Xing Jian Ji Gao Xiao De Shi Yong , Zu He Yu Ding Zhi . Zhi Chi Gao Xing Neng Bu Shu , Fu Wu Hua Bu Shu He Duan Ce Bu Shu Deng Duo Chong Bu Shu Fang Shi . Ci Wai ,Dui Yu Ge Chong Zhu Liu Ying Jian Ru Ying Wei Da GPU, Kun Lun Xin , Sheng Teng , Han Wu Ji He Hai Guang Deng ,Jin Xing Mo Xing Kai Fa Shi ,Du Ke Yi Wu Feng Qie Huan .
    • Zeng Jia Tu Xiang Yi Chang Jian Ce Suan Fa SFTPM

  • [2023-10-29] PaddleSeg 2.9Ban Ben Fa Bu !Xiang Xi Fa Ban Xin Xi Qing Can Kao Release Note.
    • Zeng Jia Dui Duo Biao Qian Fen Ge Multi-label segmentation,Ti Gong Shu Ju Zhuan Huan Dai Ma Ji Jie Guo Ke Shi Hua ,Shi Xian Dui Yi Xi Lie Yu Yi Fen Ge Mo Xing De Duo Biao Qian Fen Ge Zhi Chi .
    • Fa Bu Qing Liang Shi Jue Da Mo Xing MobileSAM,Shi Xian Geng Kuai Su De SAMTui Li .
    • Zhi Chi Liang Hua Zheng Liu Xun Lian Ya Suo Gong Neng Quant Aware Distillation Training Compression,Dui PP-LiteSeg, PP-MobileSeg, OCRNet, SegFormer-B0Zeng Jia Liang Hua Xun Lian Ya Suo Gong Neng ,Ti Sheng Tui Li Su Du .

Jian Jie

PaddleSegShi Ji Yu Fei Jiang PaddlePaddleDe Duan Dao Duan Tu Xiang Fen Ge Tao Jian ,Nei Zhi 45+Mo Xing Suan Fa Ji 140+Yu Xun Lian Mo Xing ,Zhi Chi Pei Zhi Hua Qu Dong He APIDiao Yong Kai Fa Fang Shi ,Da Tong Shu Ju Biao Zhu , Mo Xing Kai Fa , Xun Lian , Ya Suo , Bu Shu De Quan Liu Cheng ,Ti Gong Yu Yi Fen Ge , Jiao Hu Shi Fen Ge , Matting, Quan Jing Fen Ge Si Da Fen Ge Neng Li ,Zhu Li Suan Fa Zai Yi Liao , Gong Ye , Yao Gan , Yu Le Deng Chang Jing Luo Di Ying Yong .

Te Xing

  • Gao Jing Du :Gen Zong Xue Zhu Jie De Qian Yan Fen Ge Ji Zhu ,Jie He Gao Jing Du Xun Lian De Gu Gan Wang Luo ,Ti Gong 45+Zhu Liu Fen Ge Wang Luo , 150+De Gao Zhi Liang Yu Xun Lian Mo Xing ,Xiao Guo You Yu Qi Ta Kai Yuan Shi Xian .

  • Gao Xing Neng :Shi Yong Duo Jin Cheng Yi Bu I/O, Duo Qia Bing Xing Xun Lian , Ping Gu Deng Jia Su Ce Lue ,Jie He Fei Jiang He Xin Kuang Jia De Xian Cun You Hua Gong Neng ,Da Fu Du Jian Shao Fen Ge Mo Xing De Xun Lian Kai Xiao ,Rang Kai Fa Zhe Geng Di Cheng Ben , Geng Gao Xiao Di Wan Cheng Tu Xiang Fen Ge Xun Lian .

  • Mo Kuai Hua :Yuan Yu Mo Kuai Hua She Ji Si Xiang ,Jie Ou Shu Ju Zhun Bei , Fen Ge Mo Xing , Gu Gan Wang Luo , Sun Shi Han Shu Deng Bu Tong Zu Jian ,Kai Fa Zhe Ke Yi Ji Yu Shi Ji Ying Yong Chang Jing Chu Fa ,Zu Zhuang Duo Yang Hua De Pei Zhi ,Man Zu Bu Tong Xing Neng He Jing Du De Yao Qiu .

  • Quan Liu Cheng :Da Tong Shu Ju Biao Zhu , Mo Xing Kai Fa , Mo Xing Xun Lian , Mo Xing Ya Suo , Mo Xing Bu Shu Quan Liu Cheng ,Jing Guo Ye Wu Luo Di De Yan Zheng ,Rang Kai Fa Zhe Wan Cheng Yi Zhan Shi Kai Fa Gong Zuo .

Chan Pin Ju Zhen

Mo Xing Zu Jian Te Se An Li
Gu Gan Wang Luo
Sun Shi Han Shu
Ping Gu Zhi Biao
  • mIoU
  • Accuracy
  • Kappa
  • Dice
  • AUC_ROC
Zhi Chi Shu Ju Ji
Shu Ju Zeng Qiang
  • Flipping
  • Resize
  • ResizeByLong
  • ResizeByShort
  • LimitLong
  • ResizeRangeScaling
  • ResizeStepScaling
  • Normalize
  • Padding
  • PaddingByAspectRatio
  • RandomPaddingCrop
  • RandomCenterCrop
  • ScalePadding
  • RandomNoise
  • RandomBlur
  • RandomRotation
  • RandomScaleAspect
  • RandomDistort
  • RandomAffine
Fen Ge Yi Qie Mo Xing
Mo Xing Xuan Xing Gong Ju
Ren Xiang Fen Ge Mo Xing
3DYi Liao Fen Ge Mo Xing
CityscapesDa Bang Mo Xing
CVPRGuan Jun Mo Xing
Ling Yu Zi Gua Ying

Chan Ye Ji Fen Ge Mo Xing Ku

Gao Jing Du Yu Yi Fen Ge Mo Xing

Gao Jing Du Mo Xing ,Fen Ge mIoUGao , Tui Li Suan Liang Da ,Gua He Bu Shu Zai Fu Wu Qi Duan GPUHe JetsonDeng She Bei .

Mo Xing Ming Cheng Gu Gan Wang Luo CityscapesJing Du mIoU(%) V100 TRTTui Li Su Du (FPS) Pei Zhi Wen Jian
FCN HRNet_W18 78.97 24.43 yml
FCN HRNet_W48 80.70 10.16 yml
DeepLabV3 ResNet50_OS8 79.90 4.56 yml
DeepLabV3 ResNet101_OS8 80.85 3.2 yml
DeepLabV3 ResNet50_OS8 80.36 6.58 yml
DeepLabV3 ResNet101_OS8 81.10 3.94 yml
OCRNet HRNet_w18 80.67 13.26 yml
OCRNet HRNet_w48 82.15 6.17 yml
CCNet ResNet101_OS8 80.95 3.24 yml

Ce Shi Tiao Jian :

  • V100Shang Ce Su Tiao Jian :Zhen Dui Nvidia GPU V100,Shi Yong PaddleInferenceYu Ce Ku De Python API,Kai Qi TensorRTJia Su ,Shu Ju Lei Xing Shi FP32,Shu Ru Tu Xiang Wei Du Shi 1x3x1024x2048.
Qing Liang Ji Yu Yi Fen Ge Mo Xing

Qing Liang Ji Mo Xing ,Fen Ge mIoUZhong Deng , Tui Li Suan Liang Zhong Deng ,Ke Yi Bu Shu Zai Fu Wu Qi Duan GPU, Fu Wu Qi Duan X86 CPUHe Yi Dong Duan ARM CPU.

Mo Xing Ming Cheng Gu Gan Wang Luo CityscapesJing Du mIoU(%) V100 TRTTui Li Su Du (FPS) Xiao Long 855Tui Li Su Du (FPS) Pei Zhi Wen Jian
PP-LiteSeg STDC1 77.04 69.82 17.22 yml
PP-LiteSeg STDC2 79.04 54.53 11.75 yml
BiSeNetV1 - 75.19 14.67 1.53 yml
BiSeNetV2 - 73.19 61.83 13.67 yml
STDCSeg STDC1 74.74 62.24 14.51 yml
STDCSeg STDC2 77.60 51.15 10.95 yml
DDRNet_23 - 79.85 42.64 7.68 yml
HarDNet - 79.03 30.3 5.44 yml
SFNet ResNet18_OS8 78.72 10.72 - yml

Ce Shi Tiao Jian :

  • V100Shang Ce Su Tiao Jian :Zhen Dui Nvidia GPU V100,Shi Yong PaddleInferenceYu Ce Ku De Python API,Kai Qi TensorRTJia Su ,Shu Ju Lei Xing Shi FP32,Shu Ru Tu Xiang Wei Du Shi 1x3x1024x2048.
  • Xiao Long 855Shang Ce Su Tiao Jian :Zhen Dui Xiao Mi 9Shou Ji ,Shi Yong PaddleLiteYu Ce Ku De CPP API,ARMV8Bian Yi ,Dan Xian Cheng ,Shu Ru Tu Xiang Wei Du Shi 1x3x256x256.
Chao Qing Liang Ji Yu Yi Fen Ge Mo Xing

Chao Qing Liang Ji Mo Xing ,Fen Ge mIoUYi Ban , Tui Li Suan Liang Di ,Gua He Bu Shu Zai Fu Wu Qi Duan X86 CPUHe Yi Dong Duan ARM CPU.

Mo Xing Ming Cheng Gu Gan Wang Luo ADE20KJing Du mIoU(%) Xiao Long 855Tui Li Yan Shi (ms) Can Shu Liang (M) Pei Zhi Wen Jian
TopFormer-Base TopTransformer-Base 38.28 480.6 5.13 config
PP-MobileSeg-Base StrideFormer-Base 41.57 265.5 5.62 config
TopFormer-Tiny TopTransformer-Tiny 32.46 490.3 1.41 config
PP-MobileSeg-Tiny StrideFormer-Tiny 36.39 215.3 1.61 config

Ce Shi Tiao Jian :

  • Zhen Dui Xiao Mi 9Shou Ji ,Shi Yong PaddleLiteYu Ce Ku De CPP API,ARMV8Bian Yi ,Dan Xian Cheng ,Shu Ru Tu Xiang Wei Du Shi 1x3x512x512. Ce Shi Mo Xing Zai Dai You Zui Hou Yi Ge argmaxSuan Zi De Tiao Jian Xia Jin Xing Ce Shi .
Mo Xing Ming Cheng Gu Gan Wang Luo CityscapesJing Du mIoU(%) V100 TRTTui Li Su Du (FPS) Xiao Long 855Tui Li Su Du (FPS) Pei Zhi Wen Jian
MobileSeg MobileNetV2 73.94 67.57 27.01 yml
MobileSeg MobileNetV3 73.47 67.39 32.90 yml
MobileSeg Lite_HRNet_18 70.75 10.5 13.05 yml
MobileSeg ShuffleNetV2_x1_0 69.46 37.09 39.61 yml
MobileSeg GhostNet_x1_0 71.88 35.58 38.74 yml

Ce Shi Tiao Jian :

  • V100Shang Ce Su Tiao Jian :Zhen Dui Nvidia GPU V100,Shi Yong PaddleInferenceYu Ce Ku De Python API,Kai Qi TensorRTJia Su ,Shu Ju Lei Xing Shi FP32,Shu Ru Tu Xiang Wei Du Shi 1x3x1024x2048.
  • Xiao Long 855Shang Ce Su Tiao Jian :Zhen Dui Xiao Mi 9Shou Ji ,Shi Yong PaddleLiteYu Ce Ku De CPP API,ARMV8Bian Yi ,Dan Xian Cheng ,Shu Ru Tu Xiang Wei Du Shi 1x3x256x256.

Shi Yong Jiao Cheng

Ru Men Jiao Cheng

Ji Chu Jiao Cheng

Jin Jie Jiao Cheng

Huan Ying Gong Xian

Te Se Neng Li

Chan Ye Shi Jian Fan Li

Geng Duo Fan Li Xiang Mu Ke Can Kao :{Tu Xiang Fen Ge Jing Dian Xiang Mu Ji } Yong PaddleSegNeng Zuo Shi Yao ?

Xu Ke Zheng Shu

Ben Xiang Mu De Fa Bu Shou Apache 2.0 licenseXu Ke Ren Zheng .

She Qu Gong Xian

  • Fei Chang Gan Xie jm12138Gong Xian U2-NetMo Xing .
  • Fei Chang Gan Xie zjhellofss(Fu Xin Xin )Gong Xian Attention U-NetMo Xing ,He Dice lossSun Shi Han Shu .
  • Fei Chang Gan Xie liuguoyu666Gong Xian U-Net++Mo Xing .
  • Fei Chang Gan Xie yazheng0307 (Liu Zheng )Gong Xian Kuai Su Kai Shi Jiao Cheng Wen Dang .
  • Fei Chang Gan Xie CuberrChenGong Xian STDC (rethink BiSeNet) PointRend,He Detail AggregateSun Shi Han Shu .
  • Fei Chang Gan Xie stuartchen1949Gong Xian SegNet.
  • Fei Chang Gan Xie justld(Lang Du )Gong Xian UPerNet, DDRNet, CCNet, ESPNetV2, DMNet, ENCNet, HRNet_W48_Contrast, BiSeNetV1, FastFCN, SECrossEntropyLoss He PixelContrastCrossEntropyLoss.
  • Fei Chang Gan Xie Herman-Hu-saber(Hu Hui Ming )Can Yu Gong Xian ESPNetV2.
  • Fei Chang Gan Xie zhangjin12138Gong Xian Shu Ju Zeng Qiang Fang Fa RandomCenterCrop.
  • Fei Chang Gan Xie simuler Gong Xian ESPNetV1.
  • Fei Chang Gan Xie ETTR123(Zhang Kai ) Gong Xian ENet,PFPNNet.

Xue Zhu Yin Yong

Ru Guo Wo Men De Xiang Mu Zai Xue Zhu Shang Bang Zhu Dao Ni ,Qing Kao Lu Yi Xia Yin Yong :

@misc{liu2021paddleseg,
title={PaddleSeg: A High-Efficient Development Toolkit for Image Segmentation},
author={Yi Liu and Lutao Chu and Guowei Chen and Zewu Wu and Zeyu Chen and Baohua Lai and Yuying Hao},
year={2021},
eprint={2101.06175},
archivePrefix={arXiv},
primaryClass={cs.CV}
}

@misc{paddleseg2019,
title={PaddleSeg, End-to-end image segmentation kit based on PaddlePaddle},
author={PaddlePaddle Authors},
howpublished = {\url{https://github.com/PaddlePaddle/PaddleSeg}},
year={2019}
}

About

Easy-to-use image segmentation library with awesome pre-trained model zoo, supporting wide-range of practical tasks in Semantic Segmentation, Interactive Segmentation, Panoptic Segmentation, Image Matting, 3D Segmentation, etc.

Resources

Readme

License

Apache-2.0 license

Stars

Watchers

Forks

Releases

No releases published

Packages

Contributors

Languages

  • Python 90.8%
  • Java 2.6%
  • Shell 2.6%
  • C++ 1.8%
  • Cuda 1.0%
  • CMake 0.6%
  • Other 0.6%