Closed Deaddawn closed 3 weeks ago
Anyone encounter this problem?
Have you solved it? I met the same problem.
Have you solved it? I met the same problem.
Unfortunately no, but I suspect that the problem lies in the code of rasterize in diff submodule.
@ShijieZhou-UCLA This should be a bug, mind taking a look?
Sorry for the late reply! I believe it is fixed now, please kindly check!
It's OK now, thanks.
Hi, i run: python render.py -s data/bicycle/ -m output/bicycle_lseg_512/ -f lseg --iteration 4000 --edit_config configs/edit_deletion_1.yaml
then encounter : Reading camera 194/194 [20/09 02:46:09] │·· Loading Training Cameras [20/09 02:46:09] │·· [ INFO ] Encountered quite large input images (>1.6K pixels width), rescaling to 1.6K. │·· If this is not desired, please explicitly specify '--resolution/-r' as 1 [20/09 02:46:09] │·· Loading Test Cameras [20/09 02:47:27] │·· not skip train.... [20/09 02:47:54] │·· {'edit': {'objects': ['bench', 'tree', 'bicycle'], 'operations': 'deletion', 'targets': 'bench', 'threshold': 0.198}} [20/09 02:47:54] │·· Rendering progress: 0%| | 0/194 [00:02<?, ?it/s]│·· Traceback (most recent call last): │·· File "render.py", line 406, in │··
render_sets(model.extract(args), args.iteration, pipeline.extract(args), args.skip_train, args.skip_test, args.novel_view, │··
File "render.py", line 369, in render_sets │··
render_set(dataset.model_path, "train", scene.loaded_iter, scene.getTrainCameras(), gaussians, pipeline, background, edit_config, dataset.speedup) │··
File "render.py", line 130, in render_set │··
render_pkg = render_edit(view, gaussians, pipeline, background, text_feature, edit_dict) │··
File "/data/midjourney/fukunyin/fdu/feature-3dgs/gaussian_renderer/init.py", line 152, in render_edit │··
rendered_image, feature_map, radii = rasterizer( │··
ValueError: too many values to unpack (expected 3)
Can you take a look at this