{"paper":{"title":"HYPERPOSE: Hyperbolic Kinematic Phase-Space Attention for 3D Human Pose Estimation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"3D human pose estimation performed inside hyperbolic space preserves the skeleton's tree structure and avoids the volume distortion that Euclidean methods produce.","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Ajay Waghumbare, Ashish Musale, Upasna Singh, Vinduja Thekkath","submitted_at":"2026-05-11T07:17:32Z","abstract_excerpt":"We introduce HYPERPOSE, a novel 3D human pose estimation framework that performs spatio-temporal reasoning entirely within the Lorentz model of hyperbolic space $\\mathbb{H}^d$ to natively preserve the hierarchical tree topology of the human skeleton. Current state-of-the-art pose estimators aim to capture complex joint dynamics by relying on transformers and graph convolutional networks. Since these architectures operate exclusively in Euclidean space which fundamentally mismatches the inherent tree structure of the human body, these methods inevitably suffer from exponential volume distortion"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"HYPERPOSE achieves state-of-the-art structural and temporal coherence, significantly reducing both volume distortion and velocity error, while establishing new state-of-the-art benchmarks in overall positional accuracy.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That operating entirely within the Lorentz model of hyperbolic space will natively preserve the hierarchical tree topology of the human skeleton and avoid the exponential volume distortion that Euclidean methods suffer from. (Abstract, opening motivation paragraph)","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"HYPERPOSE performs 3D human pose estimation entirely in the Lorentz hyperbolic model using kinematic phase-space attention and Riemannian losses, reporting state-of-the-art structural coherence on Human3.6M and MPI-INF-3DHP.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"3D human pose estimation performed inside hyperbolic space preserves the skeleton's tree structure and avoids the volume distortion that Euclidean methods produce.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"7b03a2b706bc1247dfe8f22d51a285bef912d1049c3be9af4517f63b29e3da5b"},"source":{"id":"2605.10100","kind":"arxiv","version":2},"verdict":{"id":"b1daa544-208d-4758-b67a-5618626f716a","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T17:40:58.091304Z","strongest_claim":"HYPERPOSE achieves state-of-the-art structural and temporal coherence, significantly reducing both volume distortion and velocity error, while establishing new state-of-the-art benchmarks in overall positional accuracy.","one_line_summary":"HYPERPOSE performs 3D human pose estimation entirely in the Lorentz hyperbolic model using kinematic phase-space attention and Riemannian losses, reporting state-of-the-art structural coherence on Human3.6M and MPI-INF-3DHP.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That operating entirely within the Lorentz model of hyperbolic space will natively preserve the hierarchical tree topology of the human skeleton and avoid the exponential volume distortion that Euclidean methods suffer from. (Abstract, opening motivation paragraph)","pith_extraction_headline":"3D human pose estimation performed inside hyperbolic space preserves the skeleton's tree structure and avoids the volume distortion that Euclidean methods produce."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.10100/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-19T15:41:00.171264Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T12:01:17.796345Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T09:41:04.113038Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"ea3c786e03a4fc83510dbc71d8ce4c687512038b286a3995b31afd34f97c810c"},"references":{"count":29,"sample":[{"doi":"","year":2014,"title":"Catalin Ionescu, Dragos Papava, Vlad Olaru, and Cristian Sminchisescu. Human3.6M: Large scale datasets and predictive methods for 3D human sensing in natural environments.IEEE Transactions on Pattern ","work_id":"3704cbd5-c6a1-4544-84fc-8b97b89b2935","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2018,"title":"Cascaded pyramid network for multi-person pose estimation","work_id":"f23a3ec2-dba8-40e7-a340-3b4fa0867e4a","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2017,"title":"Julieta Martinez, Rayat Hossain, Javier Romero, and James J. Little. A simple yet effective baseline for 3D human pose estimation. InProceedings of the IEEE International Conference on Computer Vision","work_id":"4c7819e6-7fea-4775-b8fb-61494c29b063","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2019,"title":"3D human pose estimation = 2D pose estimation + matching","work_id":"de22409e-ed14-4a74-97ad-682267e18b92","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2021,"title":"3D human pose estimation with spatial and temporal transformers","work_id":"c97af412-a38d-4003-9774-950c5032f8c0","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":29,"snapshot_sha256":"bf601e63841a781e48e487ee6db88e92e3e45ab3d31cecb2a429166d34fc564d","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"79c42d215b86279a36a79271b96a0625cdf84b5b87bb32a196b2044ff825b1df"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}