To the Point: Efficient 3D Object Detection in the Range Image with Graph Convolution Kernels
This work addresses efficient and accurate 3D object detection for robotics applications, representing an incremental improvement with specific gains in pedestrian detection.
The paper tackles 3D object detection for robotics by learning directly from range images, using a 2D convolutional network with custom kernels that incorporate 3D geometry. It achieves competitive performance on the Waymo Open Dataset, improving pedestrian detection AP from 69.7% to 75.5% and offering efficiency with a model that requires 180 times fewer FLOPS and parameters than PointPillars.
3D object detection is vital for many robotics applications. For tasks where a 2D perspective range image exists, we propose to learn a 3D representation directly from this range image view. To this end, we designed a 2D convolutional network architecture that carries the 3D spherical coordinates of each pixel throughout the network. Its layers can consume any arbitrary convolution kernel in place of the default inner product kernel and exploit the underlying local geometry around each pixel. We outline four such kernels: a dense kernel according to the bag-of-words paradigm, and three graph kernels inspired by recent graph neural network advances: the Transformer, the PointNet, and the Edge Convolution. We also explore cross-modality fusion with the camera image, facilitated by operating in the perspective range image view. Our method performs competitively on the Waymo Open Dataset and improves the state-of-the-art AP for pedestrian detection from 69.7% to 75.5%. It is also efficient in that our smallest model, which still outperforms the popular PointPillars in quality, requires 180 times fewer FLOPS and model parameters