Yu Sun
Yu Sun
您好，目前的simple-romp是支持基础的tracking的，可以提取到动作序列。
@bedman367 如果你可以确定只想要图像中估计的尺度最大的人的姿态的话，可以根据估计的cam信息 https://github.com/Arthur151/ROMP/blob/e4613fd564cd632ac531228b94f949eaed76345e/romp/predict/base_predictor.py#L71 cam是3-dim的，第0个就是scale信息，可以据此获得尺度最大人的index，然后把这个人的结果提取出来。 具体代码写法可以参考： https://github.com/Arthur151/ROMP/blob/e4613fd564cd632ac531228b94f949eaed76345e/simple_romp/romp/main.py#L130
太客气了，如果有其他需要讨论的，请单独开issue，我差点漏掉了这个。
question about the difference between the intrinsics and extrinsics calibrated and estimated by ROMP
Hi. You can use our estimated 2D keypoints and 3D keypoints to re-calcualte the 3D translation of people via PnP. The reason of failure is that the 3D translation we...
question about the difference between the intrinsics and extrinsics calibrated and estimated by ROMP
Sorry, this feature is not supported by the current version, but will be supported by the next version. Best
WOW, hi, Vivian, that's a lot issues.
It does bring faster inference speed, althought not as significant as I supposed.
Yes， BEV works in the same sampling way as you described, sampling feature vecters at center and then predict the final results. ROMP is designed to have the simplest architecture...
1.Yes, BEV runs in a sampling way like that. You know that, BEV needs to integrate the depth encoding vector to be more discriminative in depth. So sampling the 2D...
About the dataset, here is the guidance: https://github.com/Arthur151/ROMP/blob/master/docs/dataset.md