bominn

icon indicating a website link https://www.linkedin.com/in/pominhsu75baa615b/ icon indicating an email [email protected]

icon company National Taiwan Univrersity, GICE icon location Taipei, Taiwan icon location MS student now, my research focus on machine learning and computer vision

Results 3 issues of bominn

細節錯誤

1 comment

hi 感謝對 flatbuffer encode 的詳細解說， 雖然沒寫過 go 但還是大致了解了 encode 的原理了 FlatBuffers 之 Encode 內序列化 table 的部分 ` 最后一步需要修正 sword 对象头部的 offset，修改成距离 vtable 的 offset。由于当前 vtable 在低地址，所以 sword 对象在它的右边，offset 为正数，offset =...

Small mistake in solver.py ?

1 comment

Hi, I think line 309 in solver.py sholud be `Y=mus`

About shared offset fields time complexity

1 comment

Hi, in Dense-RepPoints it use normal conv for reppoints_pts_refine_conv: https://github.com/justimyhxu/Dense-RepPoints/blob/master/mmdet/models/anchor_heads/dense_reppoints_head.py#L180 On the other hand, RepPoints use deformable conv for reppoints_pts_refine_conv: https://github.com/microsoft/RepPoints/blob/master/src/reppoints_head/reppoints_head.py#L144 Is that using normal conv rather than deformable conv...