Skip to content

PrimitiveNet: Primitive Instance Segmentation with Local Primitive Embedding under Adversarial Metric (ICCV 2021)

License

Notifications You must be signed in to change notification settings

hjwdzh/PrimitiveNet

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

PrimitiveNet

Source code for the paper:

Jingwei Huang, Yanfeng Zhang, Mingwei Sun. [PrimitiveNet: Primitive Instance Segmentation with Local Primitive Embedding under Adversarial Metric], ICCV 2021 .

PrimitiveNet Teaser

Compile

git submodule update --init --recursive
sh scripts/compile.sh

Download data and checkpoints

Follow the comments in download.sh to download the data and checkpoints.

sh scripts/download.sh

Evaluation on ABC

sh scripts/evaluate.sh

Predicted results on test set are visualized in src/results/visualize. Original network predictions are saved in src/results/predictions.

After all test set predictions are generated, an evaluation for mSegIOU/mLabelIOU/APs will be executed and final results will be saved at src/results/statistics.

Train ABC from scratch

sh scripts/train.sh

Logs and trained models will be saved at src/results/checkpoint.

Execute on a large scene

sh scripts/test_scene.sh

The segmented large scene is stored at src/results/visualize/final.obj.

About

PrimitiveNet: Primitive Instance Segmentation with Local Primitive Embedding under Adversarial Metric (ICCV 2021)

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published