Wei Li

Hi, @1702609 I'm facing this problem too. Have you solved this problem? Thanks a lot!

Hi, Yuying @geyuying Thanks for your awesome work! The results are really amazing！ Could you tell me your enviroment when testing CP-VTON+? I got a FID **10.186522** for PF-AFN, which...

https://github.com/jin-s13/COCO-WholeBody

https://google.github.io/mediapipe/

HPRNet: Hierarchical Point Regression for Whole-Body Human Pose Estimation https://github.com/nerminsamet/HPRNet

Yes, it's also a truncated SDF.

Hi, Rana @ranahanocka You can choose to use MC or EMC by changing the `USE_MARCHING_CUBES` macro in [this line](https://github.com/VVingerfly/surfRecon/blob/ef21ebe371695c4d0c179ed2d40b1a92dbbcee18/src/test/main.cpp#L91). Best Regards

Hi, @Hellomodo I met the same problem, have solved this problem? B.R.

H3WB: Human3.6M 3D WholeBody Dataset and Benchmark https://github.com/wholebody3d/wholebody3d

Hi, @csyhping, to produce sharp features with emc, you need to use _**enhanced distance field representation**_ rather than the simple volume representation, see [original emc paper](https://www.graphics.rwth-aachen.de/media/papers/feature1.pdf) for reference.