Search papers, labs, and topics across Lattice.
This paper introduces Diff-KD, a knowledge distillation framework that uses diffusion models to refine corrupted sensor data in multi-agent collaborative perception. Diff-KD employs Progressive Knowledge Distillation (PKD) to restore global semantics from corrupted local features using a conditional diffusion process, and Adaptive Gated Fusion (AGF) to weight neighbor contributions based on ego-vehicle reliability. Experiments on OPV2V and DAIR-V2X datasets demonstrate state-of-the-art performance in detection accuracy and calibration robustness under various corruption types.
Diffusion models can actively repair corrupted sensor data in collaborative perception, leading to significantly improved detection accuracy and robustness.
Multi-agent collaborative perception enables autonomous systems to overcome individual sensing limits through collective intelligence. However, real-world sensor and communication corruptions severely undermine this advantage. Crucially, existing approaches treat corruptions as static perturbations or passively conform to corrupted inputs, failing to actively recover the underlying clean semantics. To address this limitation, we introduce Diff-KD, a framework that integrates diffusion-based generative refinement into teacher-student knowledge distillation for robust collaborative perception. Diff-KD features two core components: (i) Progressive Knowledge Distillation (PKD), which treats local feature restoration as a conditional diffusion process to recover global semantics from corrupted observations; and (ii) Adaptive Gated Fusion (AGF), which dynamically weights neighbors based on ego reliability during fusion. Evaluated on OPV2V and DAIR-V2X under seven corruption types, Diff-KD achieves state-of-the-art performance in both detection accuracy and calibration robustness.
