From 245c4b502dc497f2a3d6dac4640184ae9d0b0539 Mon Sep 17 00:00:00 2001 From: crosstyan Date: Thu, 6 Mar 2025 12:08:49 +0800 Subject: [PATCH] docs: Improve readability of section 3.1 in Chen et al. paper - Reformatted text for better clarity and structure - Preserved original content and technical details - Enhanced visual separation of key points - Maintained consistent markdown formatting --- ...e_Estimation_at_Over_100_CVPR_2020_paper.md | 18 ++++++++++++++++-- 1 file changed, 16 insertions(+), 2 deletions(-) diff --git a/paper/Chen_Cross-View_Tracking_for_Multi-Human_3D_Pose_Estimation_at_Over_100_CVPR_2020_paper.md b/paper/Chen_Cross-View_Tracking_for_Multi-Human_3D_Pose_Estimation_at_Over_100_CVPR_2020_paper.md index 07f7458..ffcb71f 100644 --- a/paper/Chen_Cross-View_Tracking_for_Multi-Human_3D_Pose_Estimation_at_Over_100_CVPR_2020_paper.md +++ b/paper/Chen_Cross-View_Tracking_for_Multi-Human_3D_Pose_Estimation_at_Over_100_CVPR_2020_paper.md @@ -89,9 +89,23 @@ Given an unknown number of people interacting with each other in the scene cover With iterative processing, the overall computational cost increases only linearly as the number of cameras increases, and the strict synchronization between cameras is no longer required, making the solution have the potential to be applied to large-scale camera systems. Such a modification is straightforward, but not that easy to achieve, as the crossview association is generally ambiguous, especially when only one view is observed at one time. Another challenge, in this case, is to reconstruct 3D poses from different cameras when these cameras are not strictly synchronized. -To solve the problems, we construct our framework from two components: 1) cross-view tracking for body joint association, and 2) incremental 3D pose reconstruction for unsynchronized frames. Given a frame from a particular camera, the task of tracking is to associate the detected 2D human bodies with tracked targets. Here, we represent the targets in 3-space using historically estimated 3D poses. The cross-view association is therefore performed between 2D joints and 3D poses in 3-space, as detailed in Section 3.2. Subsequently, based on the association results, each 2D human body is assigned to a target or labeled as unmatched. +To solve the problems, we construct our framework from two components: -The 3D pose of each target is incrementally updated when combining the newly observed and previously retained 2D joints. Since these joints are from different times, conventional reconstruction method such as triangulation [16] is prone to inaccurate 3D locations. To deal with the unsynchronized frames, we present our incremental triangulation algorithm in Section 3.3. +1. cross-view tracking for body joint association +2. incremental 3D pose reconstruction for unsynchronized frames. + +Given a frame from a particular camera, the task of tracking is to associate the +detected 2D human bodies with tracked targets. Here, we represent the targets in +3-space using historically estimated 3D poses. The cross-view association is +therefore performed between 2D joints and 3D poses in 3-space, as detailed in +Section 3.2. Subsequently, based on the association results, each 2D human body +is assigned to a target or labeled as unmatched. + +The 3D pose of each target is incrementally updated when combining the newly +observed and previously retained 2D joints. Since these joints are from +different times, conventional reconstruction method such as triangulation [16] +is prone to inaccurate 3D locations. To deal with the unsynchronized frames, we +present our incremental triangulation algorithm in Section 3.3. ## 3.2. Cross­view tracking with geometric affinity