Error while visualising results

sonukiller commented 10 months ago

To see the prediction results for S3DIS, I kept --show and --task="lidar_det" while running test.py, but I am getting the following error:

TypeError: Visualizer.add_datasample() got an unexpected keyword argument 'vis_task' in <mmdet3d.engine.hooks.visualization_hook.Det3DVisualizationHook object at 0x7fecccbe2680>

Do you know the solution?

filaPro commented 10 months ago

As I understand mmdetection3d doesn't support instance or panoptic segmentation visualization. You can inherit from Det3DLocalVisualizer and colorize point clouds by yourself.

sonukiller commented 10 months ago

According to this, I have added below in the config file.

vis_backends = [dict(type='LocalVisBackend')]
visualizer = dict(
    type='Det3DLocalVisualizer', vis_backends=vis_backends, name='visualizer')

Then I am getting

Traceback (most recent call last):
  File "/workspace/tools/test.py", line 148, in <module>
    main()
  File "/workspace/tools/test.py", line 144, in main
    runner.test()
  File "/opt/conda/lib/python3.10/site-packages/mmengine/runner/runner.py", line 1767, in test
    metrics = self.test_loop.run()  # type: ignore
  File "/opt/conda/lib/python3.10/site-packages/mmengine/runner/loops.py", line 435, in run
    self.run_iter(idx, data_batch)
  File "/opt/conda/lib/python3.10/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context
    return func(*args, **kwargs)
  File "/opt/conda/lib/python3.10/site-packages/mmengine/runner/loops.py", line 456, in run_iter
    self.runner.call_hook(
  File "/opt/conda/lib/python3.10/site-packages/mmengine/runner/runner.py", line 1783, in call_hook
    getattr(hook, fn_name)(self, **kwargs)
  File "/opt/conda/lib/python3.10/site-packages/mmdet3d/engine/hooks/visualization_hook.py", line 207, in after_test_iter
    self._visualizer.add_datasample(
  File "/opt/conda/lib/python3.10/site-packages/mmengine/dist/utils.py", line 366, in wrapper
    return func(*args, **kwargs)
  File "/opt/conda/lib/python3.10/site-packages/mmdet3d/visualization/local_visualizer.py", line 891, in add_datasample
    self._draw_pts_sem_seg(data_input['points'],
  File "/opt/conda/lib/python3.10/site-packages/mmdet3d/visualization/local_visualizer.py", line 733, in _draw_pts_sem_seg
    pts_sem_seg = tensor2ndarray(pts_seg.pts_semantic_mask)
AttributeError: 'PointData' object has no attribute 'pts_semantic_mask'

Do you know the solution?

filaPro commented 10 months ago

I mean, you need to inherit from Det3DLocalVisualizer and a little bit modify its code for instance segmentation task.

filaPro / oneformer3d

Error while visualising results #14