Yifan Lu

That's abnormal. What's your `tensorflow` and `waymo-open-dataset` version? Some previous version can not read the map data ``` tensorflow==2.11.0 waymo-open-dataset-tf-2-11-0==1.6.1 ```

This is where `map.pkl` generate. Maybe you can put a breakpoint here and debug. And when running `python process_waymo_script.py --waymo_data_dir=../data/waymo_tfrecords/1.4.2 --nerf_data_dir=../data/waymo_multi_view`, my output is ``` 2024-05-15 16:23:16.783479: I tensorflow/core/platform/cpu_feature_guard.cc:193] This...

This is because your world coordinate is not aligning with the road surface. In Waymo's data process, we use the vehicle coordinate (red circle in the figure) from the first...

If you have a PC with display, you can remove the `-b` option in `blender -b --python blender_utils/main_multicar.py -- config/1346_multi_car_demo.yaml`. It will reproduce all the steps with opening a Blender...

What is the coordinate of your point cloud in the figure? If they are in your camera coordinate, the z value at the bottom of the cone is usually not...

The output depth may be (x^2 + y^2 + z^2) ^ 0.5 In other words, the `depth` measures the absolute distance between points and camera origin. And this causes the...

I think my guess is correct. We enable the 'z-pass' [here](https://github.com/yifanlu0227/ChatSim/blob/29a48696dc1b2e166b3a8c26093c9880d3dde845/chatsim/foreground/Blender/utils/blender_utils/render/render.py#L55) and it will create an output `Depth` socket for Render Layers Node. The saved depth.exr is the absolute distance...

Of course. You know your camera ray's direction vector. You can easily obtain the z-axis value

It's possible that OpenAI has changed their model so that some of the responses from GPT have changed from previous behavior. You can try to modify the prompt, for example...

You can also set a low `temperature` in the API call, which will make the response robust and lessen unexpected behaviors once you have a nice prompt. Also, check hyperparameters...