Keypoint detection AI Models

Models listed

50

Reviewed and live in this category

Worth weighing up

What goes in and out, licence terms, and whether you can host it yourself

How we treat data

Published specs only — we don't estimate numbers a vendor hasn't stated

Keypoint detection models are a source-backed task collection in this directory. The current editorial inventory contains 25 mapped model records. Task membership is a discovery signal, not a performance ranking, deployment recommendation, or proof that every checkpoint accepts the same inputs and returns the same outputs.

Start with the model list, then open each record that matches your data, operating constraints, and intended workflow. Check the primary source, documented modalities, license, release information, and implementation notes before using a model in production.

This collection keeps unsupported pricing, availability, and benchmark claims blank. Each page remains outside the public index until an editor approves its content and related model records.

Keypoint detection model list

Scan the table below, then open any model for pricing, limits and the full write-up.

50 models
ModelCapabilityContextInput $/1MOutput $/1MSpeed
akore/rtmw-l-256x192
akore
10.0
amirmmahdavikia/pelvitect-kp
amirmmahdavikia
10.0
ETH-CVG/lightglue_disk
ETH-CVG
10.0
ETH-CVG/lightglue_superpoint
ETH-CVG
10.0
facebook/sapiens2-pose-0.4b
facebook
10.0
facebook/sapiens2-pose-0.8b
facebook
10.0
facebook/sapiens2-pose-1b
facebook
10.0
facebook/sapiens2-pose-5b
facebook
10.0
facebook/sapiens-pose-0.3b-torchscript
facebook
10.0
facebook/sapiens-pose-0.6b
facebook
10.0
facebook/sapiens-pose-0.6b-torchscript
facebook
10.0
facebook/sapiens-pose-1b-torchscript
facebook
10.0
facebook/sapiens-pose-bbox-detector
facebook
10.0
image-matching-models/eloftr
image-matching-models
10.0
magic-leap-community/superpoint
magic-leap-community
10.0
nielsr/vitpose-base-simple
nielsr
10.0
onnx-community/vitpose-base-simple
onnx-community
10.0
qualcomm/Facial-Landmark-Detection
qualcomm
10.0
qualcomm/HRNetPose
qualcomm
10.0
qualcomm/MediaPipe-Pose-Estimation
qualcomm
10.0
qualcomm/Posenet-Mobilenet
qualcomm
10.0
SebasJanampa/DETRPose_M_COCO
SebasJanampa
10.0
SebasJanampa/DETRPose_X_COCO
SebasJanampa
10.0
SebasJanampa/DETRPose_X_CROWDPOSE
SebasJanampa
10.0
stanfordmimi/synthpose-vitpose-base-hf
stanfordmimi
10.0
stanfordmimi/synthpose-vitpose-huge-hf
stanfordmimi
10.0
usyd-community/vitpose-base
usyd-community
10.0
usyd-community/vitpose-base-coco-aic-mpii
usyd-community
10.0
usyd-community/vitpose-base-simple
usyd-community
10.0
usyd-community/vitpose-plus-base
usyd-community
10.0
usyd-community/vitpose-plus-huge
usyd-community
10.0
usyd-community/vitpose-plus-large
usyd-community
10.0
usyd-community/vitpose-plus-small
usyd-community
10.0
vismatch/aliked-lightglue
vismatch
10.0
vismatch/disk-lightglue
vismatch
10.0
vismatch/lisrd
vismatch
10.0
vismatch/loftr
vismatch
10.0
vismatch/matchanything-eloftr
vismatch
10.0
vismatch/minima
vismatch
10.0
vismatch/orb-nn
vismatch
10.0
vismatch/rdd
vismatch
10.0
vismatch/rdd-lightglue
vismatch
10.0
vismatch/roma
vismatch
10.0
vismatch/sift-lightglue
vismatch
10.0
vismatch/superpoint-lightglue
vismatch
10.0
vismatch/tiny-roma
vismatch
10.0
vismatch/xfeat
vismatch
10.0
vismatch/xfeat-lightglue
vismatch
10.0
vismatch/xfeat-star
vismatch
10.0
vrg-prague/BBoxMaskPose
vrg-prague
10.0

50 models · click a column to sort

About the capability score: a 0–100 figure SyncDev calculates from each vendor's published specifications — context window, reasoning support, input modalities, tool calling, maximum output and how recently the model shipped. It measures breadth of capability, not benchmark performance, so a higher-scoring model is not automatically the better choice for your task.

Keypoint detection model questions

It groups source-backed directory records for research and comparison. Review each model page and its primary documentation because collection membership alone does not prove quality, availability, or suitability.