-
UPDATE: June 11, 2021
**Is your feature request related to a problem? Please describe.**
Currently we use mobilnetv2 with a 300x300 input tensor (image) by default for object detection.
It stru…
-
Hi there, I am quite new to these. How can I use VITPose to run inference on my image? Do I use mmpose?
Kind regards
-
Hello @pmeier ,
I hope this issue finds you well.
I've seen you actively contribute to the torchvision/detection project, so I wanted to reach for help.
For my bachelor thesis I need to chang…
-
Hi @lisiyao21 ,
Thank you very much for your valuable code!
I set "rotmat: True", Use SMPL, but get the error as follow:
-
1.Can faster methods be used for feature extraction, and currently, a single batch of feature extraction can only be done at 15fps?
2.Can we obtain 3D joint positions in the Integrate features stag…
-
Consider the `Color` component as applied to a `Mesh3D` archetype. Does the `Color` specify the color of faces, edges, or vertices?
To support this, we could introduced _tagged_ components, allowin…
emilk updated
3 weeks ago
-
Hello, I have been struggling with fixing an error of size mismatch. I created 2d humaneva gt and 3d humaneva npz file. I think the data contributes to model_pos_train and model_pos variables in run.…
-
Hello~ I am recently testing the generalization ability of Human3.6M-trained models on the MPI-INF-3DHP dataset, but I always failed to get the results as the paper demonstrated. Could you share your …
-
(Debugged using the iOS demo app at https://github.com/tucan9389/PoseEstimation-CoreML)
The output dimension of the pretrained CPM model when converted to CoreML is 96x96; however, the resolution o…
-
### Description:
Hi,
Here is facade on top of `daisy` algorithm to make it work like `ORB`.
As result the skimage matcher and plot helper can be reused.
Maybe the match measure have no real …