在运行 -use-ba 时, qvec 问题总是存在

参数: {'scene_dir': '/content/vggt/SCENE_DIR/', 'seed': 42, 'use_ba': True, 'max_reproj_error': 8.0, 'shared_camera': False, 'camera_type': 'SIMPLE_PINHOLE', 'vis_thresh': 0.2, 'query_frame_num': 8, 'max_query_pts': 2048, 'fine_tracking': True, 'conf_thres_value': 5.0} 将 seed 设置为: 42 使用设备: cuda 使用 dtype: torch.float16 模型加载完成 从 /content/vggt/SCENE_DIR/images 中加载了 8 张图像 使用位于 /root/.cache/torch/hub/facebookresearch_dinov2_main 中的缓存 /root/.cache/torch/hub/facebookresearch_dinov2_main/dinov2/layers/swiglu_ffn.py:51: UserWarning: xFormers is not available (SwiGLU) warnings.warn("xFormers is not available (SwiGLU)") /root/.cache/torch/hub/facebookresearch_dinov2_main/dinov2/layers/attention.py:33: UserWarning: xFormers is not available (Attention) warnings.warn("xFormers is not available (Attention)") /root/.cache/torch/hub/facebookresearch_dinov2_main/dinov2/layers/block.py:40: UserWarning: xFormers is not available (Block) warnings.warn("xFormers is not available (Block)") 为了加快推理速度,请考虑禁用 fine_tracking 为查询帧 0 预测轨迹 为查询帧 3 预测轨迹 为查询帧 4 预测轨迹 为查询帧 1 预测轨迹 为查询帧 7 预测轨迹 为查询帧 6 预测轨迹 为查询帧 2 预测轨迹 为查询帧 5 预测轨迹 Traceback (most recent call last): File "/content/vggt/demo_colmap.py", line 298, in demo_fn(args) File "/content/vggt/demo_colmap.py", line 173, in demo_fn reconstruction, valid_track_mask = batch_np_matrix_to_pycolmap( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/content/vggt/vggt/dependency/np_to_pycolmap.py", line 109, in batch_np_matrix_to_pycolmap pose_ba = pycolmap.Rigid3d(pycolmap.Rotation3d(qvec[i]), tvec[i]) ^^^^ NameError: name 'qvec' is not defined