mit-han-lab/bevfusion
[ICRA'23] BEVFusion: Multi-Task Multi-Sensor Fusion with Unified Bird's-Eye View Representation
PythonApache-2.0
Issues
- 6
Performance is worse while training
#582 opened by 186big - 1
Evaluation Process is stuck vol.2
#583 opened by EpicGilgamesh - 1
Missing some implementation of transfusion?
#586 opened by YueSun0609 - 3
learning rate configuration
#590 opened by konyul - 1
error raised by setup.py
#593 opened by fdy61 - 12
KeyError: "BEVFusion: 'encoders.camera.backbone.stages.0.blocks.0.attn.w_msa.relative_position_bias_table'"
#597 opened by fdy61 - 1
Why is the height of LiDAR voxelized grid 41
#600 opened by duanmushuangquan - 2
RuntimeError: CUDA error: CUBLAS_STATUS_INVALID_VALUE when calling `cublasGemmEx( handle, opa, opb, m, n, k, &falpha, a, CUDA_R_16F, lda, b, CUDA_R_16F, ldb, &fbeta, c, CUDA_R_16F, ldc, CUDA_R_32F, CUBLAS_GEMM_DFALT_TENSOR_OP)`
#602 opened by lingxe01 - 1
problems about test and visualize
#604 opened by 18lalala - 10
The problem with the test
#607 opened by Ange1ika - 1
- 1
- 1
config for camera+lidar with centerhead
#618 opened by zyh321 - 2
- 2
About the test accuracy not reaching
#622 opened by npuxdy - 1
What changes are needed to modify the detection range to 100 meters before and after?
#623 opened by Tamako668 - 1
Consultation on Windows training file modification
#624 opened by huidabao - 1
- 1
Results for 1 stage
#626 opened by konyul - 1
- 1
KeyError: 'radar'
#628 opened by F0sakens3-1415 - 2
- 1
- 0
Visualization from pkl
#620 opened by gerardmartin2 - 6
How to change the lidar backbone to pointpillars
#613 opened by wyf0414 - 22
- 0
- 7
mAP The difference between train and test AND what is the training strategy
#619 opened by Li-Whasaka - 2
CAN'T REPRODUCE THE RESULT IN PAPER
#594 opened by Li-Whasaka - 1
- 1
I met errors,please help
#614 opened by sunpian1 - 2
what is the lidar-only-det.pth
#611 opened by wyf0414 - 1
The training of bevfusion only modifies the POINT-CLOUD-RANGE algorithm, resulting in significant differences in effectiveness
#599 opened by zzwforever - 1
Question about how to pretrain swin-T on nuimages?
#591 opened by liyih - 1
Test time augmentation
#588 opened by skmadan - 1
run the torchpack command and error
#610 opened by Lawrence-Cheung - 2
Image BEV feature was getting worse during training
#587 opened by san9569 - 1
I can't download the pretained model now
#608 opened by hitbuyi - 1
Is there a tiny model for a small GPU memory?
#609 opened by hitbuyi - 1
fatal: unable to access 'https://github.com/mit-han-lab/bevfusion/': gnutls_handshake() failed: Error in the pull function.
#601 opened by OrangeJuice3211 - 1
- 1
How to use RTX4090 to run the code when cuda11.8 not support lower versions of torch and mmcv-full-1.4.0
#606 opened by Wangquans - 3
Reproducing code with low accuracy
#612 opened by xmutyjs - 1
Will Waymo's configuration be released?
#595 opened by MaolinGanMSU - 1
Why is this fix needed? newest release of mmdetection3d still does not do this
#598 opened by kabouzeid - 2
The mAP and NDS are lower than the offical.
#605 opened by wyf0414 - 2
Inquiry regarding use of RGB-D camera
#592 opened by Poodlee - 8
Pre-processing Error NuScenesDataset: [Errno 2] No such file or directory: 'nuscenes_infos_train.pkl'
#589 opened by anithselva - 0
How to save best checkpoint?
#585 opened by heli223 - 1
`download_pretrained.sh` fails due to 404
#584 opened by josh-wende