通过潜在扩散空间实现2D/3D步态轨迹对齐，提升跨模态识别精度。
DiffCrossGait: Trajectory-Level Alignment for 2D-3D Cross-Modal Gait Recognition via Latent Diffusion
- 在潜在扩散空间中进行轨迹级对齐，而非仅对齐最终特征
- 在SUSTech1K和FreeGait上达到当前最优性能
- 生成对齐与分类解耦，推理效率高，适合实际部署
跨模态2D-3D步态识别受限于2D轮廓与3D LiDAR范围图表示之间的固有域差异。现有方法仅对齐最终嵌入，我们提出DiffCrossGait，将跨模态匹配重新定义为身份相关潜在扩散空间中的轨迹级对齐，而非假设2D与3D观测完全等价。通过在潜在空间中使用共享高斯噪声驱动双模态，实现在生成演化过程中的连续对齐。引入三阶段对齐策略，利用不同噪声强度分别强化身份锚定、动态一致性与跨模态结构可恢复性，从而约束双模态共享去噪动态与瓶颈结构，促进模态不变的步态特征。关键在于，该框架将生成对齐与判别主干解耦，扩散机制仅作为训练目标，避免迭代去噪的计算开销，保障高推理效率。在SUSTech1K与FreeGait基准上的大量实验表明，DiffCrossGait取得最先进性能。
原文摘要 · Abstract (English)
Cross-modal 2D-3D gait recognition is impeded by inherent domain discrepancies between 2D silhouette and 3D LiDAR range-view representations. While prior methods align only final embeddings, we propose DiffCrossGait, which reformulates cross-modal matching as trajectory-level alignment in an identity-relevant latent diffusion space, rather than assuming full equivalence between 2D and 3D observations. By driving both modalities with shared Gaussian noise within a latent space, we enable continuous alignment throughout the generative evolution. We introduce a Tri-Phase Alignment Strategy that exploits varying noise intensities to enforce identity anchoring, dynamics consistency, and cross-modal structural recoverability, thereby constraining both modalities to share denoising dynamics and bottleneck structure, which promotes modality-invariant gait features. Crucially, our framework decouples generative alignment from the discriminative backbone; the diffusion mechanism serves exclusively as a training objective, ensuring high inference efficiency by eliminating the computational overhead of iterative denoising. Extensive experiments on the SUSTech1K and FreeGait benchmarks demonstrate that DiffCrossGait achieves state-of-the-art performance.
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。