-
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…
-
### What is the feature?
我成功调用了单webcam进行3d human pose detection。请问目前支持多webcam，通过视差、三角测量等技术来获取三维姿态嘛？如果可以的话，需要怎么做呢
### Any other context?
_No response_
-
## TODO
- [x] MOT Tracking
- [x] MOT eval by Pose detection
- [x] MOT eval by Pose tracking
- [x] MOT eval by mrcnn detection
- [x] MOT eval by default detection
- [x] Pose est…
-
我要生成1024乘1024的图片,用control net导入图片生成pose图,那么导入图片的尺寸有什么要求?节点的参数怎么设置会影响清晰度?
-
- Have I written custom code:
No
- OS Platform and Distribution (e.g., Linux Ubuntu 16.04):
Android 14,API 34 OneUI6.0
- Mobile device (e.g. iPhone 8, Pixel 2, Samsung Galaxy) if the issue happens…
-
Dear @DavidBoja
Thanks so much for your contributions.
Is there a way to use Mediapipe, to use it for your measurement SMPL anthropometry ? If that's possible what needs to be changed and how ?
-
Thank you for sharing such an impressive project! I have a question about the calculation of Euler angles from 2D keypoints in the projected image, specifically in the function „calculate_euler_angles…
-
I would like to ask for the dataset used for the object detection task. I missed the registration deadline, so I could not download the dataset. I would really appreciate it if you could share the dat…
-
PrefetchDataZMQ for more information.
File "E:/company/CodeCubao/Detect_DL/Pose-key-points-detection-master/lib/train.py", line 30, in
trainer = Trainer(Model(), cfg)
File "E:\company\Code…
-
In the run tracking demo on videos, the paper mentions that the BERT-style transformer model(pose_transformer_v2) enables future predictions and amodal completion of missing detections within the same…