Home▸AI & Machine Learning▸Two-Stream Action Recognition in 2D: Spatial vs Temporal Fusion

Two-Stream Action Recognition in 2D: Spatial vs Temporal Fusion

A 2D stick-figure skeleton performs four actions while a spatial (pose-shape) stream and a temporal (joint-velocity) stream each vote on what it's doing, late-fused into one prediction — drag to pan, scroll to zoom, and watch a live signal strip show why the motion stream degrades first under noise.

AI & Machine Learning2DModerate60 FPS📱 Mobile-adapted⇄ 3D version

Real video action-recognition systems — from the original two-stream network to I3D and SlowFast — classify what's happening by combining two very different signals: an appearance stream that reads a single frame's shape, and a motion stream that reads how things move between frames. This 2D canvas simulation drives a stick-figure skeleton through four actions (walk, wave, jump, squat) using the same forward-kinematics equations as the 3D original, extracts a live pose-shape descriptor and a live joint-velocity descriptor, scores each against per-action templates with cosine similarity and softmax, and late-fuses the two predictions with an adjustable weight. A pannable, zoomable stage plus a live scrolling signal strip make visible why the motion stream is the more fragile of the two under sensor noise.

⚙ Under the hood

action recognitionvideo analysistwo-stream networkoptical flowkinematicsAI

2D · HTML5 Canvas 2D · 60 FPS target · runs fully client-side, no install