Holistic Landmark

Search FilterSearch indexed videos with natural language.Search indexed videos with natural language.CleanupRemove content not needed in the video.Remove content not needed in the video.SegmentationBox objects and extract task plus low-level action labels.Box objects and extract task plus low-level action labels.Hand PoseEstimate hand landmarks for finger skeleton overlay.Estimate hand landmarks for finger skeleton overlay.Depth MapEstimate depth maps for colorized video overlay.Estimate depth maps for colorized video overlay.DiffusionGenerate styled video variations using state-of-the-art video-to-video models.Generate styled video variations using state-of-the-art video-to-video models.Image to SimulatorReconstruct physics-ready MuJoCo simulations from multi-view images.Reconstruct physics-ready MuJoCo simulations from multi-view images.AnalyzeCreate data points to learn better behavior.Create data points to learn better behavior.Full Body PoseTrack 33 body landmarks in image and hip-centered 3D coordinates, and preview a skeleton over the video.Track 33 body landmarks in image and hip-centered 3D coordinates, and preview a skeleton over the video.Face LandmarkTrack 478 face landmarks plus 52 expression scores, and preview a face mesh over the video.Track 478 face landmarks plus 52 expression scores, and preview a face mesh over the video.Gesture DetectionLabel each hand with a gesture and confidence, and preview the name on the hand.Label each hand with a gesture and confidence, and preview the name on the hand.Holistic LandmarkTrack 543 pose, face, and hand landmarks together (33 + 468 + 42), and preview them on the video.Track 543 pose, face, and hand landmarks together (33 + 468 + 42), and preview them on the video.Mask SegmentationTrace each detected object’s outline as a polygon, separate from box segmentation.Trace each detected object’s outline as a polygon, separate from box segmentation.

Supported holistic landmark models

ModelsDatasetsExamplesNotesPlayground