Pose Estimation Models Compared: MediaPipe, MoveNet, YOLO and More

For a single-user, on-device fitness app, MediaPipe/BlazePose is the usual starting point: it predicts 33 body landmarks and offers monocular 3D "world" landmarks on a normal phone. For lightweight browser or mobile 2D tracking, MoveNet (17 keypoints) gives you a speed-vs-accuracy dial — Lightning for speed, Thunder for accuracy. For multi-person scenes (gyms, group classes), YOLO-pose or OpenPose handle many people at once. The one trade-off that decides most builds: single-person on-device (private, cheap, simpler) vs. multi-person (heavier, and with real licensing strings attached). If you only understand one thing before picking, understand the license — OpenPose is non-commercial and Ultralytics YOLO is AGPL-3.0.

New to the concept itself? Start with what pose estimation is; this page is about choosing among the actual models.

The models you actually choose among

Keypoint counts below are stable facts you can rely on. Everything about speed is described as positioning, not a number — fps and accuracy vary wildly by device, resolution, and delegate, so any benchmark must be verified on your own target hardware (as of 2026, verify).

• MediaPipe / BlazePose (Google). Predicts 33 landmarks (a superset of the 17 COCO keypoints, adding face, hand, and foot points). It is designed for on-device, real-time use and outputs both 2D image-coordinate landmarks and 3D world landmarks in meters, plus per-landmark visibility. It is single-person focused. License: Apache-2.0 (verify current terms for bundled model assets).
• MoveNet — Lightning vs Thunder (TensorFlow / Google). Detects 17 keypoints. Shipped in two variants: Lightning is tuned for latency-critical/speed use, Thunder for higher accuracy. Single-person oriented; commonly used for browser/web and lightweight mobile. License: Apache-2.0.
• YOLO-pose / Ultralytics (YOLO11-pose). Default model uses 17 keypoints stored as (x, y, visibility) triplets, COCO-Pose trained. It is multi-person capable, detecting people and their keypoints in one pass. Model sizes (n/s/m/l/x) trade speed for accuracy. License: AGPL-3.0 (strong copyleft) with a paid Enterprise License to bypass it — see the licensing note below.
• PoseNet (legacy, TensorFlow.js). 17 keypoints, now deprecated in favor of the newer pose-detection package (which serves MoveNet and BlazePose). Older and lighter, generally lower accuracy. Prefer MoveNet or BlazePose for new projects. License: Apache-2.0.
• OpenPose (CMU, research). A bottom-up multi-person system using Part Affinity Fields to group keypoints into people. Configurable keypoint sets (COCO-18 / BODY_25 variants — verify current config). Heavy and GPU-oriented, strong for research and multi-person, less suited to lightweight on-device. License: non-commercial / academic research from CMU — commercial use needs a separate license.

Comparison table

Positioning, not fabricated benchmarks. Confirm licenses and any speed claim before you ship.

The licensing trap (read before you commit)

Two of these models carry license terms that can block a commercial product, and they are easy to miss because the code is free to download:

• Ultralytics YOLO is AGPL-3.0. AGPL is a strong reciprocal (copyleft) license: if you distribute or even offer a network service built on AGPL code, its terms can require you to release your own source under the same license. Ultralytics sells a paid Enterprise License specifically to lift that obligation. Treat "which license" as a real legal decision, not a footnote — verify the current terms and talk to counsel before building a closed-source product on it.
• OpenPose is non-commercial / academic. CMU's license permits research use; commercial use requires a separate agreement. Do not ship it in a paid or commercial app without clearing licensing first.

The Apache-2.0 options (MediaPipe/BlazePose, MoveNet, PoseNet) are the most permissive here, though you should still verify terms for any bundled model assets.

How to decide — by use case

There is no single winner; the right model follows your scene and constraints:

• Single user, on-device, needs joint angles (and optionally 3D): MediaPipe/BlazePose. The 33 landmarks and monocular 3D world landmarks make it the default for a private, phone-based rep-counting or form-feedback app.
• Single user, pure speed, browser or lightweight mobile, 2D only: MoveNet — Lightning when responsiveness beats precision, Thunder when you can afford more compute for a cleaner pose.
• Multiple people at once (group class, gym floor, sports scene): YOLO-pose (mind AGPL-3.0) or OpenPose (mind the non-commercial license). Multi-person also needs a separate cross-frame tracking layer to keep identities stable.
• New build: avoid PoseNet — it is legacy; pick MoveNet or BlazePose instead.

If you would rather not choose and maintain a model at all, a managed motion SDK bundles pose plus rep/form logic — see AI workout tracking APIs for the buy option.

The honest limits

• A normal RGB phone camera is enough. No depth sensor or LiDAR is required for 2D or monocular 3D pose.
• Monocular 3D depth is estimated, not measured. BlazePose world landmarks are hip-relative and inferred from a single camera; depth is unreliable for occluded or extremity joints and degrades on out-of-plane movement. It is not sensor-grade — treat the depth axis as lower-confidence than x/y, smooth it, and validate. More reliable 3D needs multi-camera rigs or depth sensors.
• Benchmarks are condition-dependent. Leaderboard fps and accuracy assume specific hardware, lighting, and framing. Do not present a lab number as your app's real-world result; test on your own cameras, lighting, distances, and exercises.
• Form feedback is a coaching aid, not medical or physical-therapy advice. Monocular angle error, occlusion, and depth ambiguity mean it can be wrong; it cannot judge pain or injury. Recommend a professional for anything injury-related.
• Streaming raw video to a server is privacy-sensitive. Keeping inference on-device (video never leaves the phone) is the strongest privacy posture.

Before you ship

Pick by use case, then verify: confirm the current license (especially OpenPose non-commercial and YOLO AGPL-3.0), re-check any fps/accuracy claim on your own target hardware, and validate monocular 3D on real users rather than trusting a lab figure. When you are ready to wire it into a camera feed, see the camera pose tracking guide for implementation.