{"arxiv_id":"2503.19625","incoming":{"papers":0,"sentences":0,"groups":[],"unlabelled_sentences":0},"outgoing":{"papers":4,"sentences":9,"groups":[{"kind":"method_or_tool","label":"Uses the method or tool of","papers":2,"sentences":3,"quotes":[{"arxiv_id":"2410.11831","title":"CoTracker3: Simpler and Better Point Tracking by Pseudo-Labelling Real Videos","url":"/paper/2410.11831","section":"Relative Object Pose Estimation via Point Tracking","sentence":"Our method adopts a similar strategy but replaces the keypoint detection and matching component with a state-of-the-art Track Any Point (TAP) model [ref] .","intent_p":0.99,"contests_p":null},{"arxiv_id":"2408.00714","title":"SAM 2: Segment Anything in Images and Videos","url":"/paper/2408.00714","section":"Unseen Object Pose Estimation","sentence":"To ensure fairness, each sequence is paired with a specific CAD model and uniform segmentations produced by SAM2 [ref] .","intent_p":0.98,"contests_p":null}]},{"kind":"background","label":"Cites as background","papers":2,"sentences":2,"quotes":[{"arxiv_id":"2502.04293","title":"GCE-Pose: Global Context Enhancement for Category-level Object Pose Estimation","url":"/paper/2502.04293","section":"Category-level Object Pose Estimation","sentence":"Furthermore, some methods directly predict object pose, employing either enhanced geometric feature extractors [ref] or a combination of geometric and semantic priors [ref] .","intent_p":0.97,"contests_p":null},{"arxiv_id":"2405.07801","title":null,"url":"/paper/2405.07801","section":"Object Pose Tracking","sentence":"Unlike single-frame object pose estimation methods, object pose tracking leverage temporal information across video sequences to estimate object poses [ref] .","intent_p":0.95,"contests_p":null}]}],"unlabelled_sentences":4},"claims":[],"thresholds":{"intent_p":0.8,"supports_p":0.95,"contests_p":0.7},"method":"Citing sentences come from each paper's LaTeX source; every \\cite marks the sentence and the works it cites. Pith's classifier typed each sentence. A label is shown at intent probability 0.8 or more (0.95 for support), a contest at 0.7 or more. Measured accuracy against two blind model families is at https://pith.science/accuracy. The quoted sentences are the citing authors' own.","held_up":null,"_links":{"html":"https://pith.science/paper/2503.19625"}}