frametop-hands-models

An IRON export of facebook/sapiens2-pose-1b for AMD Ryzen AI NPUs: the compiled NPU kernels (.xclbin + instruction streams) and the packed weights the taconite-sapiens2 Rust runtime replays, over XRT or directly over the amdxdna driver. The kernels are compiled for NPU2 (AIE2P: Strix Point, Strix Halo, Krackan) and will not load on NPU1 (Phoenix, Hawk Point). The larger Sapiens2-Pose: 40 layers of 1536 (the 0.4b's 24 of 1024), the same 308 keypoints. Top-down: give each person's box (COCO x,y,w,h, from a person detector); the keypoints come back in image coordinates with their heatmap scores. Needs taconite-sapiens2 0.1.2 or later (bundle format 2). sapiens2 check on a Ryzen AI 9 HX 370…

Track any point through a video. Give it a video and a set of points; it returns where each point is on every frame, whether it can see it, and how sure it is. It runs at 0.0065 s/frame and returns a calibrated per-frame confidence alongside every track. It is LocoTrack-B with cross-track attention added — the tracks attend to each other, so a point that is hidden can be inferred from the neighbours that are still visible — then fine-tuned for occlusion — on MOVi-E, and in the newer checkpoint on Meta's Apache-2.0 Kubric release. This repository is gated. Click Request access; the maintainer approves manually. Once The LocoTrack baseline weights are not mirrored here; they come from the…

Model · Keypoint detection

vitpose-plus-base

University of Sydney

ViTPose: Simple Vision Transformer Baselines for Human Pose Estimation and ViTPose+: Vision Transformer Foundation Model for Generic Body Pose Estimation. It obtains 81.1 AP on MS COCO Keypoint test-dev set. Although no specific domain knowledge is considered in the design, plain vision transformers have shown excellent performance in visual recognition tasks. However, little effort has been made to reveal the potential of such simple structures for pose estimation tasks. In this paper, we show the surprisingly good capabilities of plain vision transformers for pose estimation from various aspects, namely simplicity in model structure, scalability in model size, flexibility in training…