-
hi
I am on xavier NX jetpack 4.4DP (so tensorrt 7.1)
I have build successfully (I think) all the steps for Modenet until this one:
`LD_LIBRARY_PATH=$(pwd)/onnx-tensorrt/build onnx-tensorrt/build/…
-
@ZHKKKe Thanks for sharing your inference code and the pre-trained model.
I took your pre-trained PyTorch model and was able to build a TensorRT engine from it. I shared my TensorRT implementatio…
-
Hi, thank you very much for your great work!
I tried the Self-Supervised sub-objective consistency (SOC) adaptation iteration of MODNet but i get an error and i cannot figure out why. I'd appreciate …
-
##### System information (version)
- OpenVINO=> 2021.4
- Operating System / Platform => Ubuntu 18.04.5 LTS
- Python Version => 3.9.5
- Compiler => GCC 9.3.0
- Problem classification => Network Bi…
-
view_np = matte_np * frame_np + (1 - matte_np) * frame_dy
只这步就得25~28ms，我们就拿26来算，也就38帧，是怎么得到一秒钟63帧的，而且分辨率好低，如果用HD或者4K的视频，
吓死个人好吧
_, _, matte_tensor = modnet(frame_tensor, True)
…
-
In your MODNet class, you should not set `self.backbone` if backbone is not used anywhere else in that class. You pass the backbone to LRBranch and LRBranch set `self.backbone` again. So the same mobi…
-
This is happening in both single player and multiplayer games, and with both the 64-bit and regular clients.
Throws this error in the log:
[01:14:55] [1/WARN] [tML]: Silently Caught Exception:
Sys…
-
When I export the TorchScript version of MODNet, error occurs:
```
Traceback (most recent call last):
File "/usr/local/lib/python3.6/runpy.py", line 193, in _run_module_as_main
"__main__", …
-
-
你好 想请问关于在MODnet模型中所使用的训练数据集，
目前得知因为有Licence问题，所以训练集不会发布
也知道你们是自己搜集人物影像，并利用photoshop标注。
我想请教的是:
1. 此训练集中半身跟全身影像的比例?
2. 从issue上面得知，你们是搜集3000张前景，与30张(512*512)的背景，以不同比例的方式将前景合成到背景中，
最后共获得90000张图片去训练…