Shiyi Lan

icon indicating a website link https://voidrank.github.io icon indicating an email [email protected]

icon company NVIDIA Research icon location Sunnyvale, CA icon location Full-stack Research Scientist. Interests: Visual Recognition; Auto-labeling; Foundation Models; Autonomous Driving; CUDA

Results 23 comments of Shiyi Lan

Issues with IoU Computation in The Eval Code

Hi @chengchunhsu Thanks for the confirmation. Will be great if you can also pay attention to the second issue I mentioned. There may be another bug of how ignored pixels...

Issues with IoU Computation in The Eval Code

To be more specific, the *ignored pixels* means those edge pixels marked as 255th category, which should be ignored in computing IoU. However, I found that the evaluation code doesn't...

Have you attempted to perform weakly-supervised semantic edge extraction?

Hi @Yancccccc The only concern is that, in DiscoBox or [MAL](https://github.com/NVlabs/mask-auto-labeler), the training data is fed in a way that only one object is in the center of the image....