1. https://hanlab.mit.edu/projects/bevfusion, see the video

Multi-sensor fusion

Why BEV?

Untitled

What is BEV and how?

‣

Untitled

好像没讲depth的supervision，不清楚有没有直接关于depth 的loss，但是相关讨论可以参考：‣

以前也有人做，本文we diagnose and lift key efficiency bottlenecks in the view transformation with optimized BEV pooling, reducing latency by more than 40×.

Untitled

我们用，不关心起实现。

References

  1. BEVFusion: Multi-Task Multi-Sensor Fusion with Unified Bird's-Eye View Representation, ICRA 23
  2. https://github.com/ADLab-AutoDrive/BEVFusion, nips22
  3. https://github.com/vasgaowei/BEV-Perception#lidar-camera
  4. https://github.com/HuangJunJie2017/BEVDet