Part of GloSL, a set of continuous sign language translation corpora in one format. annotations.csv, card.json, poses/ /.pose. card.json holds the full provenance. annotations.csv columns (same in every GloSL dataset; empty when not applicable): id, split, text, gloss, glosstype, signer, start, end, fps, nframes, videoref, posefile, videofile. rtmpose-wholebody (Uni-Sign), 133 points (BODY133 23, FACE133 68, LEFTHAND133 21, RIGHTHAND133 21), 2D. Coordinates: pixels of the frame the pose refers to (header width x height, depth 0); z raw. Confidence: RTMPose score; missing point = (0,0) with confidence 0. Frame rate: native, per file (also in annotations.csv). Normalization (glosl-norm-v1) is…
Independent publisher
Pedro Dal Bianco
pedroodb
Models in Library0
Datasets in Library1
Models on Hugging Face—
Followers—
Compare the published evidence
This profile connects the 1 model and dataset record shown below to their upstream cards and sources. The library counts describe SAVRN's coverage, rather than this publisher's entire output.
Published tasks: Translation.
Reported licenses: cc-by-nc-4.0. Open a record to review its specific license and conditions.
Compare model specifications, source dates, and hardware estimates on the individual records. Publisher identity verification is separate from a model's quality, deployment readiness, or suitability for your work.