On Denoising Walking Videos for Gait Recognition
This work addresses the challenge of improving gait recognition accuracy for applications like surveillance or biometrics by reducing noise from clothing and other distractions, representing an incremental advance over existing denoising methods.
The paper tackles the problem of removing identity-irrelevant cues like clothing in walking videos for gait recognition by proposing DenoisingGait, a method that uses diffusion models and a geometry-driven feature matching module to create a flow-like gait representation, achieving state-of-the-art performance on datasets such as CCPG, CASIA-B*, and SUSTech1K.
To capture individual gait patterns, excluding identity-irrelevant cues in walking videos, such as clothing texture and color, remains a persistent challenge for vision-based gait recognition. Traditional silhouette- and pose-based methods, though theoretically effective at removing such distractions, often fall short of high accuracy due to their sparse and less informative inputs. Emerging end-to-end methods address this by directly denoising RGB videos using human priors. Building on this trend, we propose DenoisingGait, a novel gait denoising method. Inspired by the philosophy that "what I cannot create, I do not understand", we turn to generative diffusion models, uncovering how they partially filter out irrelevant factors for gait understanding. Additionally, we introduce a geometry-driven Feature Matching module, which, combined with background removal via human silhouettes, condenses the multi-channel diffusion features at each foreground pixel into a two-channel direction vector. Specifically, the proposed within- and cross-frame matching respectively capture the local vectorized structures of gait appearance and motion, producing a novel flow-like gait representation termed Gait Feature Field, which further reduces residual noise in diffusion features. Experiments on the CCPG, CASIA-B*, and SUSTech1K datasets demonstrate that DenoisingGait achieves a new SoTA performance in most cases for both within- and cross-domain evaluations. Code is available at https://github.com/ShiqiYu/OpenGait.