Model Zoo¶
Common settings¶
- We use distributed training.
- For fair comparison with other codebases, we report the GPU memory as the maximum value of
torch.cuda.max_memory_allocated()
for all 8 GPUs. Note that this value is usually less than whatnvidia-smi
shows. - We report the inference time as the total time of network forwarding and post-processing, excluding the data loading time. Results are obtained with the script benchmark.py which computes the average time on 2000 images.
Baselines¶
SECOND¶
Please refer to SECOND for details. We provide SECOND baselines on KITTI and Waymo datasets.
PointPillars¶
Please refer to PointPillars for details. We provide pointpillars baselines on KITTI, nuScenes, Lyft, and Waymo datasets.
VoteNet¶
Please refer to VoteNet for details. We provide VoteNet baselines on ScanNet and SUNRGBD datasets.
Dynamic Voxelization¶
Please refer to Dynamic Voxelization for details.
RegNetX¶
Please refer to RegNet for details. We provide pointpillars baselines with RegNetX backbones on nuScenes and Lyft datasets currently.
nuImages¶
We also support baseline models on nuImages dataset. Please refer to nuImages for details. We report Mask R-CNN, Cascade Mask R-CNN and HTC results currently.
CenterPoint¶
Please refer to CenterPoint for details.
SSN¶
Please refer to SSN for details. We provide pointpillars with shape-aware grouping heads used in SSN on the nuScenes and Lyft datasets currently.
ImVoteNet¶
Please refer to ImVoteNet for details. We provide ImVoteNet baselines on SUNRGBD dataset.
FCOS3D¶
Please refer to FCOS3D for details. We provide FCOS3D baselines on the nuScenes dataset currently.
PointNet++¶
Please refer to PointNet++ for details. We provide PointNet++ baselines on ScanNet and S3DIS datasets.
Group-Free-3D¶
Please refer to Group-Free-3D for details. We provide Group-Free-3D baselines on ScanNet dataset.
ImVoxelNet¶
Please refer to ImVoxelNet for details. We provide ImVoxelNet baselines on KITTI dataset.
Model Zoo Statistics¶
- Number of papers: 20
- ALGORITHM: 17
- BACKBONE: 1
- DATASET: 1
- OTHERS: 1
- Number of checkpoints: 84
- [ALGORITHM] 3DSSD: Point-based 3D Single Stage Object Detector (1 ckpts)
- [ALGORITHM] Center-based 3D Object Detection and Tracking (6 ckpts)
- [ALGORITHM] Dynamic Voxelization (3 ckpts)
- [ALGORITHM] FCOS3D: Fully Convolutional One-Stage Monocular 3D Object Detection (2 ckpts)
- [OTHERS] Mixed Precision Training (4 ckpts)
- [ALGORITHM] FreeAnchor for 3D Object Detection (8 ckpts)
- [ALGORITHM] Group-Free 3D Object Detection via Transformers (4 ckpts)
- [ALGORITHM] H3DNet: 3D Object Detection Using Hybrid Geometric Primitives (1 ckpts)
- [ALGORITHM] ImVoteNet: Boosting 3D Object Detection in Point Clouds with Image Votes (2 ckpts)
- [ALGORITHM] ImVoxelNet: Image to Voxels Projection for Monocular and Multi-View General-Purpose 3D Object Detection (1 ckpts)
- [ALGORITHM] MVX-Net: Multimodal VoxelNet for 3D Object Detection (1 ckpts)
- [DATASET] NuImages Results (14 ckpts)
- [ALGORITHM] PAConv: Position Adaptive Convolution with Dynamic Kernel Assembling on Point Clouds (1 ckpts)
- [ALGORITHM] From Points to Parts: 3D Object Detection from Point Cloud with Part-aware and Part-aggregation Network (2 ckpts)
- [ALGORITHM] PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space (6 ckpts)
- [ALGORITHM] PointPillars: Fast Encoders for Object Detection from Point Clouds (8 ckpts)
- [BACKBONE] Designing Network Design Spaces (9 ckpts)
- [ALGORITHM] Second: Sparsely embedded convolutional detection (2 ckpts)
- [ALGORITHM] SSN: Shape Signature Networks for Multi-class Object Detection from Point Clouds (7 ckpts)
- [ALGORITHM] Deep Hough Voting for 3D Object Detection in Point Clouds (2 ckpts)