Seohee Park comments

Results 13 comments of


                                            Seohee Park

Opencv loads BGR images thus lowering detection accuracy

thank you!

detect one single image using tf-cpn

I have tested that code. However, I found that the code was wrong because I guessed the pose without detecting the bounding box.

Why an extra 6 classes?

Hi ! This is not a matter of adding six classes. Just change the name of the class. [coco.names] -> [origin_coco.names] motobike -> motorcycle pottedplant -> potted plant diningtable ->...

SMPL-H model hits einsum error with hand poses

I solved it by updating the file from old SMPL file to updated SMPL file.

Hello. Having a problem with your code. Bug in SMPL model:

I solved it by updating the file from SMPLX_NEUTRAL.npz(108.8MB) to SMPLX_NEUTRAL_2020.npz (167.3MB)

Attempt to call an undefined function glGenVertexArrays, check for bool(glGenVertexArrays) before calling

Hi. In my case, the problem was caused by the A100 GPU not supporting WDDM. This was resolved by running it on an RTX series GPU.

ONNX infer

@csukuangfj Good news! Are there any plans for a Korean Onnx version?

ONNX infer

@csukuangfj I will analyze the Chinese + English version and attempt to perform inference on the Korean ONNX version. Thank you!

In this code, you can get the bert value through the **get_bert** function. Bert calls a different torch model for each language, and there is only a Python implementation. https://github.com/myshell-ai/MeloTTS/blob/144a0980fac43411153209cf08a1998e3c161e10/melo/utils.py#L22

Questions related to MeloTTS

In your code, there is a part where bert and ja_bert are entered as model inputs in ModelWrapper. https://github.com/k2-fsa/sherpa-onnx/blob/963aaba82b01a425ae8dcf0fdcff6b073a45686f/scripts/melo-tts/export-onnx.py#L172 So, even though I specified input_names as below when exporting to...