shihwesley

shihwesley / SamPlaysBaseball

Public

Pitcher mechanics analyzer: single-camera video → 3D biomechanical analysis using SAM 3D Body. Built for MLB player development.

28
2
100% credibility
Found Apr 13, 2026 at 28 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

SamPlaysBaseball is a web app for baseball coaches and analysts that uses AI to compare pitching deliveries from game videos, visualize differences in 3D, and generate diagnostic reports.

How It Works

1
🔍 Discover the tool

You hear about a simple web app that turns baseball game clips into 3D motion analysis for spotting pitching issues like tipping.

2
💻 Open the dashboard

Head to the website and see a clean interface ready for your questions about any pitcher's game.

3
💬 Ask in plain English

Type something like 'Compare his sliders to fastballs from the 6th inning' — no special format needed.

4
⏳ Watch it work

The tool grabs the game clips, analyzes the motions, and builds 3D models automatically.

5
👀 View side-by-side 3D

Scrub through the deliveries with ghost overlays to see exact differences frame by frame.

6
📋 Read the insights

Get a plain-English report naming the key mechanical changes and what they mean for performance.

✅ Coach with confidence

Share the visuals and report with your pitcher to fix issues like tipping before the next game.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 28 to 28 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is SamPlaysBaseball?

SamPlaysBaseball turns single-camera baseball video—like MLB broadcast clips or phone footage—into 3D biomechanical analysis of pitcher mechanics. Upload a clip or query Statcast data by natural language ("Compare sliders to fastballs in Ohtani's last outing"), and it delivers phase-aligned 3D side-by-side views, joint-angle charts, and diagnostic reports on arm slot, timing, and tipping signals. Built in Python with MLX for fast local inference on Apple Silicon, it's designed for post-game pitcher mechanics analysis without specialized hardware.

Why is it gaining traction?

Unlike multi-camera systems like KinaTrax that lock you into stadium installs, this handles any video source with ~55mm joint accuracy—good enough for relative comparisons like delivery differences or lefty pitcher mechanics tweaks. The query-driven dashboard with React Three Fiber 3D scrubbing, ghost overlays, and LLM narratives makes spotting pro pitcher mechanics issues (submarine slots, fatigue drift) instant and visual. Zero cloud costs and Apple-native speed hook analysts tired of manual frame-by-frame review.

Who should use this?

MLB or MiLB pitching coaches confirming live suspicions like "he's tipping his slider" via post-game footage. Player development staff analyzing softball pitcher mechanics or MLB pitcher mechanics drills from bullpens, showcases, or archives. Scouts or analysts building baselines for biomechanical body analysis without buying mocap rigs.

Verdict

Solid prototype for pitcher mechanics analyzer in player development—try the demo on Ohtani data to see 3D tipping confirmation shine. At 28 stars and 1.0% credibility, it's early but actively developed with strong docs and validation notes; integrate via FastAPI endpoints once you validate against your clips.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.