Diffusion-based 3D Object Detection with Random Boxes

TL;DR

This paper introduces Diff3Det, a diffusion model-based approach for 3D object detection that generates proposals by reversing a noise process, offering a promising alternative to traditional anchor-based methods.

Contribution

It adapts diffusion models for 3D object proposal generation, replacing heuristic anchors with a generative noise-reversal process for improved detection.

Findings

Achieves promising results on the KITTI benchmark

Outperforms some classical anchor-based methods

Demonstrates the effectiveness of diffusion models in 3D detection

Abstract

3D object detection is an essential task for achieving autonomous driving. Existing anchor-based detection methods rely on empirical heuristics setting of anchors, which makes the algorithms lack elegance. In recent years, we have witnessed the rise of several generative models, among which diffusion models show great potential for learning the transformation of two distributions. Our proposed Diff3Det migrates the diffusion model to proposal generation for 3D object detection by considering the detection boxes as generative targets. During training, the object boxes diffuse from the ground truth boxes to the Gaussian distribution, and the decoder learns to reverse this noise process. In the inference stage, the model progressively refines a set of random boxes to the prediction results. We provide detailed experiments on the KITTI benchmark and achieve promising performance compared to classical anchor-based 3D detection methods.