DeltaFlow: An Efficient Multi-frame Scene Flow Estimation Method

TL;DR

DeltaFlow introduces a lightweight multi-frame scene flow estimation framework that efficiently captures temporal motion cues, improves accuracy with novel loss functions, and outperforms existing methods in speed and error reduction.

Contribution

It proposes a novel $ riangle$ scheme for efficient multi-frame feature extraction and introduces Category-Balanced and Instance Consistency Losses to enhance flow accuracy.

Findings

Achieves up to 22% lower error on benchmark datasets.

Runs twice as fast as comparable multi-frame methods.

Demonstrates strong cross-domain generalization.

Abstract

Previous dominant methods for scene flow estimation focus mainly on input from two consecutive frames, neglecting valuable information in the temporal domain. While recent trends shift towards multi-frame reasoning, they suffer from rapidly escalating computational costs as the number of frames grows. To leverage temporal information more efficiently, we propose DeltaFlow ($\Delta$Flow), a lightweight 3D framework that captures motion cues via a $\Delta$ scheme, extracting temporal features with minimal computational cost, regardless of the number of frames. Additionally, scene flow estimation faces challenges such as imbalanced object class distributions and motion inconsistency. To tackle these issues, we introduce a Category-Balanced Loss to enhance learning across underrepresented classes and an Instance Consistency Loss to enforce coherent object motion, improving flow accuracy. Extensive evaluations on the Argoverse 2, Waymo and nuScenes datasets show that $\Delta$Flow achieves state-of-the-art performance with up to 22% lower error and $2\times$ faster inference compared to the next-best multi-frame supervised method, while also demonstrating a strong cross-domain generalization ability. The code is open-sourced at https://github.com/Kin-Zhang/DeltaFlow along with trained model weights.