Great work! Discussion about oracle performance of GTA to Cityscapes.

lhoyer / HRDA

[ECCV22] Official Implementation of HRDA: Context-Aware High-Resolution Domain-Adaptive Semantic Segmentation

Other

233 stars 31 forks source link

Great work! Discussion about oracle performance of GTA to Cityscapes. #6

Closed xingchenzhao closed 2 years ago

xingchenzhao commented 2 years ago

Thank you for your great work!

As the paper shows, the performance of GTA to Cityscapes can achieve 73.8 mIoU which is absolutely outstanding. As HRDA is based on DAFormer structure, I checked DAFormer's oracle performance of Cityscapes which is 77.6 mIoU. As we can see, 73.8 is just slightly lower than 77.6. Therefore, does it mean that there's no much space for us to improve GTA to Cityscapes unless we improve the oracle performance?

lhoyer commented 2 years ago

Hi @xingchenzhao,

Thank you for your interest in our work!

When comparing HRDA with supervised training, the high-resolution oracle (shown in Fig. 4) should be used for a more fair comparison. Therefore, there is still decent space for improvement (73.8 vs 81.2 mIoU).

Best regards, Lukas

xingchenzhao commented 2 years ago

Thank you for your reply!