likegod
likegod
> Hi > In these 'simple model' case, It is caused by `adaptiveavgpool2d`/`adaptivemaxpool2d`, Here is the difference > > official onnx this > static o o o > dynamic-output_size=1 x...
“In this way, because use keypoint location + amodel_offset instead of 2d box center solve 3D center will be more accurate ?” I have the same problem. Yes. I think...
is it ready? > Sorry for the delay @HeroyiuWFY . I will release this with 1-2 weeks.
> @ideafold 你好，请问你用的xavier是哪一款？NX 还是 AGX? agx，32G mem, 32 emmc.
> > 例如在compute_box_3d时 这个repo的方法： float coordinates_ori[8][3] = {l/2, h/2, w/2, l/2, h/2, -w/2, -l/2, h/2, -w/2, -l/2, h/2, w/2, l/2, -h/2, w/2, l/2, -h/2, -w/2, -l/2, -h/2, -w/2, -l/2, -h/2,...