AI Rank
中文
← Open Source
foamliu

Deep-Image-Matting-PyTorch

Deep Image Matting implementation in PyTorch.

Model DevelopmentArchitecturePython
Open on GitHub
Momentum
+0stars in 24 hours0.0%
832
Stars
181
Forks
+0
This week
2
Contributors
Created 2019-06-21 · Updated 2026-09-01 · #13366 today
Top developers
foamliuwrrJasmine

foamliu/Deep-Image-Matting-PyTorch is ranked 13,366 with 832 stars.

Same-category comparison

foamliu/Deep-Image-Matting-PyTorch is ranked 13,366 among these 8 peers; smilelight/lightNLP is ranked 13,369.

Same-category comparison
NameRankMetrics
finegrain-ai/refiners13,356
Language:
Python
Stars:
833
Forks:
63
Description:
Refiners (discontinued early 2025) was a microframework on top of PyTorch with first-class citizen APIs for foundation model adaptation.
koaning/human-learn13,358
Language:
Jupyter Notebook
Stars:
833
Forks:
54
Description:
Natural Intelligence is still a pretty good idea.
tonl-dev/tonl13,361
Language:
TypeScript
Stars:
833
Forks:
36
Description:
TONL (Token-Optimized Notation Language).
wujiyang/Face_Pytorch13,362
Language:
Python
Stars:
833
Forks:
154
Description:
face recognition algorithms in pytorch framework, including arcface, cosface, sphereface and so on.
smilelight/lightNLP13,369
Language:
Python
Stars:
832
Forks:
209
somaticio/tensorflow.rb13,370
Language:
Ruby
Stars:
832
Forks:
61
Description:
tensorflow for ruby.
elki-project/elki13,373
Language:
Java
Stars:
831
Forks:
322
Description:
ELKI Data Mining Toolkit.
WangRongsheng/ChatGenTitle13,383
Language:
Python
Stars:
830
Forks:
70
Key parameters and specifications
Key parameters and specifications
ParameterValue
LanguagePython
CategoryModel Development
SubcategoryArchitecture
Stars832
Forks181
Contributors2
Created date2019-06-21
Updated date2026-09-01
Stars in 1 day0
Stars in 7 days0
Rank13,366
Previous rank13,267
Rank change-99
7-day stars rank13,904
Previous 7-day stars rank13,927
7-day stars rank change23
Total stars rank12,909
Previous total stars rank12,888
Total stars rank change-21
CountryChina
README summary

Deep Image Matting Deep Image Matting paper implementation in PyTorch. "fc6" is clumpy, over 100 millions parameters, makes the model hard to converge. I guess it is the reason why the model (paper) has to be trained stagewisely. Performance The Composition-1k testing dataset. Both erode and dialte to generate trimap. 0 Dataset Adobe Deep Image Matting Dataset Follow the instruction to contact author for the dataset. MSCOCO Go to MSCOCO to download: 2014 Train images PASCAL VOC Go to

PASCAL VOC to download: VOC challenge 2008 training/validation data The test data for the VOC2008 challenge Usage Data Pre-processing Extract training images: Train If you want to visualize during training, run in your terminal: Experimental results The Composition-1k testing dataset 1. Test: It prints out average SAD and MSE errors when finished. Download the evaluation datasets: Go to the Datasets page and download the evaluation datasets. Make sure you pick the low-resolution dataset.

Frequently asked questions

How many stars and forks are recorded?

foamliu/Deep-Image-Matting-PyTorch has 832 stars and 181 forks.

What is the repository rank?

foamliu/Deep-Image-Matting-PyTorch is ranked 13,366.

What language and category are recorded?

foamliu/Deep-Image-Matting-PyTorch has 832 stars, uses Python and belongs to Model Development.

Which repository is closest in rank?

foamliu/Deep-Image-Matting-PyTorch is compared with smilelight/lightNLP, ranked 13,369 with 832 stars.

Nearby in the ranking
Similar items