DeltaFlow通过轻量Δ机制高效利用多帧信息，提升场景流精度与速度。
DeltaFlow: An Efficient Multi-frame Scene Flow Estimation Method
- 用Δ方案提取时序特征，计算开销不随帧数增加而上升
- 在三个数据集上误差降低22%，推理速度提升一倍
- 适合需要高精度实时场景流的自动驾驶应用
以往主流的场景流估计方法主要依赖连续两帧输入，忽视了时间域中的宝贵信息。尽管近期趋势转向多帧推理，但随着帧数增加，计算成本迅速上升。为更高效地利用时序信息，我们提出DeltaFlow（ΔFlow）——一种轻量级3D框架，通过Δ机制以极低计算代价提取时序特征，且不受帧数影响。此外，场景流估计面临类别分布不均和运动不一致的问题。为此，我们引入类别平衡损失以增强对少数类的学习，并设计实例一致性损失以约束物体运动的一致性，从而提升流估计精度。在Argoverse 2、Waymo和nuScenes数据集上的大量实验表明，ΔFlow在保持最高性能的同时，相比次优的多帧监督方法，误差降低最多达22%，推理速度提升2倍，并展现出强跨域泛化能力。代码及训练模型已开源至https://github.com/Kin-Zhang/DeltaFlow。
原文摘要 · Abstract (English)
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 ($Δ$Flow), a lightweight 3D framework that captures motion cues via a $Δ$ 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 $Δ$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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。