We lifted 10 functions out of this paper's own repositories and ran 3 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.
| Repository | Role | Ran |
|---|---|---|
| vitae-transformer/bevsimdet | canonical | 3 of 10 |
| Function | Status | Where it lives |
|---|---|---|
| denormalize_bbox | Ran | vitae-transformer/bevsimdet/mmdet3d/core/bbox/util.py code served (permissive licence) · get_code("38e90d12ef38fe79") |
| gaussian_2d | Ran | vitae-transformer/bevsimdet/mmdet3d/core/utils/gaussian.py code served (permissive licence) · get_code("ea036153b550642a") |
| normalize_bbox | Ran | vitae-transformer/bevsimdet/mmdet3d/core/bbox/util.py code served (permissive licence) · get_code("e6ab81c2ff932920") |
| aligned_3d_nms | Not yet run | vitae-transformer/bevsimdet/mmdet3d/core/post_processing/box3d_nms.py code served (permissive licence) · get_code("0079ed3e99e4048d") |
| box_camera_to_lidar | Not yet run | vitae-transformer/bevsimdet/mmdet3d/core/bbox/box_np_ops.py code served (permissive licence) · get_code("96c8084488bd023b") |
| camera_to_lidar | Not yet run | vitae-transformer/bevsimdet/mmdet3d/core/bbox/box_np_ops.py code served (permissive licence) · get_code("9e22984820c39868") |
| circle_nms | Not yet run | vitae-transformer/bevsimdet/mmdet3d/core/post_processing/box3d_nms.py code served (permissive licence) · get_code("a8007c7a213459b2") |
| corners_nd | Not yet run | vitae-transformer/bevsimdet/mmdet3d/core/bbox/box_np_ops.py code served (permissive licence) · get_code("5b1dacd1f125c08f") |
| draw_heatmap_gaussian | Not yet run | vitae-transformer/bevsimdet/mmdet3d/core/utils/gaussian.py code served (permissive licence) · get_code("a7cde40ccc6422e2") |
| gaussian_radius | Not yet run | vitae-transformer/bevsimdet/mmdet3d/core/utils/gaussian.py code served (permissive licence) · get_code("73e8316be86a800d") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
Multi-view camera-based 3D object detection has become popular due to its low cost, but accurately inferring 3D geometry solely from camera data remains challenging and may lead to inferior performance. Although distilling precise 3D geometry knowledge from LiDAR data could help tackle this challenge, the benefits of LiDAR information could be greatly hindered by the significant modality gap between different sensory modalities. To address this issue, we propose a Simulated multi-modal Distillation (SimDistill) method by carefully crafting the model architecture and distillation strategy. Specifically, we devise multi-modal architectures for both teacher and student models, including a LiDAR-camera fusion-based teacher and a simulated fusion-based student. Owing to the ``identical'' architecture design, the student can mimic the teacher to generate multi-modal features with merely multi-view images as input, where a geometry compensation module is introduced to bridge the modality gap. Furthermore, we propose a comprehensive multi-modal distillation scheme that supports intra-modal, cross-modal, and multi-modal fusion distillation simultaneously in the Bird's-eye-view space. Incorporating them together, our SimDistill can learn better feature representations for 3D object detection while maintaining a cost-effective camera-only deployment. Extensive experiments validate the effectiveness and superiority of SimDistill over state-of-the-art methods, achieving an improvement of 4.8\% mAP and 4.1\% NDS over the baseline detector. The source code will be released at https://github.com/ViTAE-Transformer/SimDistill.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2303.16818")
get_code_for_paper("2303.16818")
have("2303.16818")
Connect an agent — have() is free.