Science for Sport Podcast

311: The Future of Weight Room Monitoring with Perch P2

Mar 16, 2026
Jordan Lucier, Senior Director of Engineering at Catapult and co-founder of Perch, builds camera-based tools using computer vision and machine learning for weight-room measurement. He discusses Perch P2’s portability and hardware upgrades, higher frame rates and 3D vision, scaling multi-user tracking, and the push from output metrics toward movement quality and integrated on-field insights.
Ask episode
AI Snips
Chapters
Transcript
Episode notes
INSIGHT

Doubling Frame Rate Closes The Inaccuracy Gap

  • P2 doubles frame rate from 30 to 60 fps and uses a new 3D camera module to shrink the 'inaccuracy gap' on explosive movements.
  • Jordan explains higher frame rate and processing power improves resilience to fast movements and measurement accuracy.
INSIGHT

Stereo Color Vision Improves Real World Robustness

  • P2 moved from infrared depth sensing to stereo color vision, improving robustness in challenging lighting like sunlight-filled weight rooms.
  • Jordan says stereo color vision enhances 3D perception and performance across varied gym lighting.
INSIGHT

Weight Room Training Is Often Siloed And Partially Tracked

  • Perch software shows many programmed movements go untracked, revealing a siloed view of gym training.
  • Jordan wants to expand tracking beyond barbells to capture more of the workout and close gaps in coverage.
Get the Snipd Podcast app to discover more snips from this episode
Get the app