Hand instance segmentation for egocentric video, made to feed HaMeR 3D hand reconstruction. RF-DETR-Seg 2XLarge (rfdetr 1.11.1) at 768 px, fine-tuned on SAM 3.1 pseudo-labels of EgoDex and in-house MCAP recordings. Code and training procedure: RF-DETR-HandGlove (handseg/docs/training.md, docs/workflow.md). Internal research use only. The training data includes EgoDex (CC-BY-NC-ND 4.0), so these weights must not be used commercially or redistributed. A commercial model has to be retrained on in-house data. Each folder: checkpoint.pth (EMA weights, rfdetr format), rfdetr-seg-2xlarge-768fp16.onnx (static 1x3x768x768, outputs dets / labels / masks), evalresults.md (+ evalresults.md: ONNX /…
Independent publisher
Daniel Hou
danielhou-anngine
Models in Library1
Datasets in Library0
Models on Hugging Face1
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: Image segmentation.
Reported licenses: other. 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.