{"paper":{"title":"FingerEye: Learning Dexterous Manipulation with Continuous Vision-Tactile Sensing","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"FingerEye combines binocular cameras and a marker-tracked compliant ring to create one continuous perception stream from vision to tactile wrench estimates.","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Lin Shao, Tianyu Qiu, Xuanye Wu, Yichen Li, Zhixuan Xu","submitted_at":"2026-04-22T15:37:34Z","abstract_excerpt":"Dexterous robotic manipulation requires perception that remains informative from pre-contact approach to contact initiation and post-contact control. We introduce FingerEye, a sensing and learning framework that strengthens robotic dexterity through continuous vision-tactile feedback throughout interaction. On the sensing side, FingerEye integrates binocular RGB cameras with a compliant contact interface to support perception both before and after contact. Before contact, the fingertip cameras provide close-range visual cues and implicit stereo for precise approach and object localization. Aft"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"FingerEye integrates binocular RGB cameras for close-range visual perception with implicit stereo depth and marker-based pose estimation on a compliant ring to serve as a proxy for contact wrench sensing, enabling a perception stream that smoothly transitions from pre-contact visual cues to post-contact tactile feedback and supporting vision-tactile imitation learning for dexterous manipulation.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That deformations captured via marker-based pose estimation on the compliant ring structure provide a sufficiently accurate and general proxy for full contact wrench (forces and torques) across diverse interaction regimes without significant limitations from material properties, calibration drift, or range constraints.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"FingerEye delivers continuous vision-tactile sensing via binocular RGB cameras and marker-tracked compliant ring deformation, supporting imitation learning policies that generalize across object variations for tasks like coin standing and syringe manipulation.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"FingerEye combines binocular cameras and a marker-tracked compliant ring to create one continuous perception stream from vision to tactile wrench estimates.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"303e46bc22f68f9eae2edc0fa9bea465a8f524ca80b5a40ee2bc5dc674043dc7"},"source":{"id":"2604.20689","kind":"arxiv","version":3},"verdict":{"id":"74e06f15-620a-4396-823a-0b939865fea7","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-09T23:40:46.817331Z","strongest_claim":"FingerEye integrates binocular RGB cameras for close-range visual perception with implicit stereo depth and marker-based pose estimation on a compliant ring to serve as a proxy for contact wrench sensing, enabling a perception stream that smoothly transitions from pre-contact visual cues to post-contact tactile feedback and supporting vision-tactile imitation learning for dexterous manipulation.","one_line_summary":"FingerEye delivers continuous vision-tactile sensing via binocular RGB cameras and marker-tracked compliant ring deformation, supporting imitation learning policies that generalize across object variations for tasks like coin standing and syringe manipulation.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That deformations captured via marker-based pose estimation on the compliant ring structure provide a sufficiently accurate and general proxy for full contact wrench (forces and torques) across diverse interaction regimes without significant limitations from material properties, calibration drift, or range constraints.","pith_extraction_headline":"FingerEye combines binocular cameras and a marker-tracked compliant ring to create one continuous perception stream from vision to tactile wrench estimates."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.20689/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-21T14:34:24.953739Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-20T01:39:40.968467Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"6780ac49d26f968e5ac74a54da49cca4c7a1512068ce352c4bf890c9f0fb37fa"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}