DCHM: Depth-Consistent Human Modeling for Multiview Detection
This work addresses multiview pedestrian detection for surveillance and robotics by improving human modeling accuracy without relying on costly human-labeled annotations, though it appears incremental as it builds on existing two-stage frameworks.
The paper tackles the problem of noisy and imprecise human modeling in multiview pedestrian detection by proposing Depth-Consistent Human Modeling (DCHM), which uses superpixel-wise Gaussian Splatting to achieve consistent depth estimation and multiview fusion, significantly reducing noise and outperforming previous state-of-the-art methods.
Multiview pedestrian detection typically involves two stages: human modeling and pedestrian localization. Human modeling represents pedestrians in 3D space by fusing multiview information, making its quality crucial for detection accuracy. However, existing methods often introduce noise and have low precision. While some approaches reduce noise by fitting on costly multiview 3D annotations, they often struggle to generalize across diverse scenes. To eliminate reliance on human-labeled annotations and accurately model humans, we propose Depth-Consistent Human Modeling (DCHM), a framework designed for consistent depth estimation and multiview fusion in global coordinates. Specifically, our proposed pipeline with superpixel-wise Gaussian Splatting achieves multiview depth consistency in sparse-view, large-scaled, and crowded scenarios, producing precise point clouds for pedestrian localization. Extensive validations demonstrate that our method significantly reduces noise during human modeling, outperforming previous state-of-the-art baselines. Additionally, to our knowledge, DCHM is the first to reconstruct pedestrians and perform multiview segmentation in such a challenging setting. Code is available on the \href{https://jiahao-ma.github.io/DCHM/}{project page}.