Once the inference is done with Depth-Anything-V2, it saved for each image a corresponding PLY (XYZ+RGB). For this, you had to give the intrinsics camera parameters.
You can then visualize the results with :
python3 viz.py path/to/the/plys/
This will open up an Open3D basic visualizer:
Open3D viz
Then use SPACE to start/pause/resume frames and press N to go to next frame.