YOLOStereo3D: A Step Back to 2D for Efficient Stereo 3D Detection
This enables efficient 3D detection for low-cost autonomous robots without LiDAR, though it is incremental as it builds on existing 2D methods.
The paper tackles the problem of computationally expensive stereo 3D object detection by proposing YOLOStereo3D, which adapts 2D detection frameworks with a lightweight stereo module, achieving performance comparable to state-of-the-art methods at over 10 fps on a single GPU.
Object detection in 3D with stereo cameras is an important problem in computer vision, and is particularly crucial in low-cost autonomous mobile robots without LiDARs. Nowadays, most of the best-performing frameworks for stereo 3D object detection are based on dense depth reconstruction from disparity estimation, making them extremely computationally expensive. To enable real-world deployments of vision detection with binocular images, we take a step back to gain insights from 2D image-based detection frameworks and enhance them with stereo features. We incorporate knowledge and the inference structure from real-time one-stage 2D/3D object detector and introduce a light-weight stereo matching module. Our proposed framework, YOLOStereo3D, is trained on one single GPU and runs at more than ten fps. It demonstrates performance comparable to state-of-the-art stereo 3D detection frameworks without usage of LiDAR data. The code will be published in https://github.com/Owen-Liuyuxuan/visualDet3D.