Remote Photoplethysmography (rPPG) detects heartbeat-induced skin color changes via camera to extract the Blood Volume Pulse (BVP) signal, which helps measure vital signs like heart and respiratory rates. Despite its popularity in health research for its user-friendly and noninvasive approach, rPPG's BVP signal quality can be compromised by factors like video compression. This research seeks to evaluate how various compression methods affect BVP signal quality, aiming to enhance rPPG's practical use. We created the ZJXU-MOTION dataset to assess video compression's impact on Blood Volume Pulse (BVP) signal quality during motion, considering activity and lighting. We compared common codecs like H.264, H.265, AV1/VP9, MJPEG, ProRes, and FFV1, analyzing how compression artifacts and bitrates influence BVP. Our results were confirmed with the UBFC-Phys dataset. Inter-frame compression can degrade BVP signal quality by adding noise. For clear signals, use intra-frame compression like H.265 with a GOP size of 1. If using inter-frame, keep GOP small and CQP stable. For static videos, H.265 CQP or VP9 VBR with GOP sizes of 1–3 work well. For low-motion videos, H.265 provides better performance. For complex motion, lossless compression methods like FFV1 outperform H.264 and H.265 in preserving BVP signal quality. Given that inter-frame compression introduces noise into BVP processing, we recommend intra-frame compression – particularly H.265 with a GOP of 1 – for optimal signal extraction. For complex motion, H.265 and FFV1 offer robust solutions, with FFV1 being especially effective in high-motion conditions. These findings are instrumental for advancing rPPG technology in health monitoring and diagnostics.