Bui Khuong Duy

@lycooool + I think calling `camera_head` and `depth_head` separately instead of the simpler `predictions = model(images)` might be because this demo doesn't require `predict_world_points`, which is invoked at [this line](https://github.com/facebookresearch/vggt/blob/main/vggt/models/vggt.py#L77)....

It was helpful for me to install a version of PaddleX that matches PaddleOCR, which means you need to run: ``` pip uninstall -y paddlex pip install paddlex==3.1.0 ```