Evaluation problem

[TShoreWind]

Hello author, how to evaluate the model code, this is very important to me, looking forward to the author's reply

[TShoreWind]

[Yzmblog]

Hi, the evaluation codes are at https://github.com/Yzmblog/MonoHuman/blob/main/run.py#L193. Please first divide the dataset, and you can try to use this command:

python run.py \
    --type eval\
    --cfg configs/monohuman/zju_mocap/387/387.yaml \

Or you can generate the images first, and refer to the eval codes to calculate the metrics.

[TShoreWind]

Thank you for taking time out of your busy schedule to reply, I don't know what is wrong or what, there is a mistake in the evaluation after dividing the data, can you share the detailed evaluation process?