{"id":"2a97c9f3-7796-43ff-9c13-5c37ee4fc589","arxiv_id":"2606.17376","paper_version":1,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":4.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":0,"one_line_summary":"A modality-adaptive contactless RR monitoring framework for heterogeneous mobile robots that combines brightness-adaptive sensor selection, keypoint-guided ROI extraction, and SQI filtering, generalizing across platforms without retuning.","lead":"The paper describes a multimodal framework using RGB, thermal, NIR, and low-light cameras on mobile robots for contactless respiratory rate monitoring with edge computing and adaptive sensor selection. This approach could support safer remote victim assessment in hazardous environments like disasters where physical contact poses risks.","discovery_kind":"new_application","skeptic_critique":{"model":"grok-4.3","headline":"No significant objection identified","rationale":"Reader's weakest_assumption correctly isolates the robustness requirement for the no-retuning claim. With full text now available and no contradictory details or missing controls apparent, the existing UNVERDICTED verdict stands; the experiments as summarized already address the load-bearing condition.","tokens_in":1798,"tokens_out":222,"duration_ms":20062,"concrete_test":"If code or supplementary material is released, re-execute the full pipeline on held-out data from one platform using only the parameters reported for the others; confirm that RR error stays within the per-modality bounds stated in the results.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim (generalization across platforms without retuning, plus modality boundaries) is directly supported by the described experiments on three heterogeneous robots under varied conditions. The keypoint-guided ROI + SQI pipeline is presented as the mechanism enabling this, and the abstract states that results demonstrate the required robustness. No internal inconsistency or unsupported leap is visible from the provided content.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.3","summary":"The paper presents a modality-adaptive contactless respiratory rate (RR) monitoring framework for heterogeneous mobile robots with onboard edge computing. It combines brightness-adaptive sensor selection across RGB, thermal, NIR, and low-light cameras, keypoint-guided chest ROI extraction, and SQI-based filtering. The framework is implemented on three robotic platforms (quadruped and wheeled) and evaluated across lighting conditions, poses, and distances; the central claim is that it generalizes across platforms without per-platform retuning while revealing modality-specific boundaries (RGB to 8 m, NIR to 6 m, thermal short range only, low-light to 8 m in darkness).","tokens_in":1843,"tokens_out":477,"duration_ms":33000,"significance":"If the generalization and boundary claims hold, the work would offer a practical engineering contribution for contactless vital-sign monitoring in hazardous search-and-rescue and triage scenarios. The explicit multimodal design and deployment on three heterogeneous platforms with edge computing are strengths that could inform future autonomous systems.","major_comments":[{"comment":"Abstract and experimental evaluation sections: the manuscript asserts that experiments demonstrate generalization without retuning and the listed modality-specific ranges, yet supplies no quantitative metrics (e.g., RR estimation error, detection success rates at each distance, or cross-platform statistical comparisons) to support these claims.","section":"Abstract and experimental evaluation sections"},{"comment":"Method sections describing keypoint-guided ROI extraction and SQI filtering: these components are presented as the mechanism enabling posture-robust performance without platform-specific retuning, but no equations, pseudocode, threshold values, or ablation results on posture/illumination changes are provided to substantiate reliability across conditions.","section":"Method sections describing keypoint-guided ROI extraction and SQI filtering"}],"minor_comments":[{"comment":"Figure captions and axis labels should explicitly state the robot platform, distance, and lighting condition for each example to aid interpretation.","section":"Figures"},{"comment":"The manuscript would benefit from a short table summarizing the three robotic platforms (locomotion type, compute hardware, camera suite) for quick reference.","section":"Platform description"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":{"model":"grok-4.3","summary":"Thank you for the referee's insightful comments. We address the major concerns below by committing to specific revisions that will strengthen the quantitative and methodological support in the manuscript.","responses":[{"response":"We agree that additional quantitative metrics are necessary to support the claims. The revised manuscript will include detailed tables with RR estimation errors (MAE and RMSE), detection success rates at varying distances for each modality, and cross-platform statistical comparisons to demonstrate generalization without retuning.","revision_made":"yes","referee_comment":"[Abstract and experimental evaluation sections] Abstract and experimental evaluation sections: the manuscript asserts that experiments demonstrate generalization without retuning and the listed modality-specific ranges, yet supplies no quantitative metrics (e.g., RR estimation error, detection success rates at each distance, or cross-platform statistical comparisons) to support these claims."},{"response":"We will expand the method sections in the revision to include the mathematical equations for keypoint-guided ROI extraction and SQI computation, pseudocode for the overall process, the specific threshold values used, and ablation results showing the impact of posture and illumination variations on performance.","revision_made":"yes","referee_comment":"[Method sections describing keypoint-guided ROI extraction and SQI filtering] Method sections describing keypoint-guided ROI extraction and SQI filtering: these components are presented as the mechanism enabling posture-robust performance without platform-specific retuning, but no equations, pseudocode, threshold values, or ablation results on posture/illumination changes are provided to substantiate reliability across conditions."}],"tokens_in":1438,"tokens_out":337,"duration_ms":27778,"standing_objections":[]},"desk_editor":{"model":"grok-4.3","letter":"The punchline is that the authors put together a brightness-adaptive camera switcher, keypoint chest ROI, and SQI filter, ran it on a quadruped and two wheeled robots with different edge hardware, and showed it works across lighting, poses, and distances up to 8 m without rewriting the code for each platform. RGB covers the widest range, NIR to 6 m, thermal only close in, and low-light works in darkness to 8 m.\n\nWhat is new is the specific combination for heterogeneous mobile robots in edge settings aimed at search-and-rescue triage. Earlier contactless RR work existed, but this one ties the pieces to robot locomotion types and onboard compute while testing the no-retuning claim.\n\nIt does well on the practical side. The experiments actually move the robots around real subjects under varied conditions and report usable modality boundaries. That kind of hardware-grounded comparison is helpful for anyone who needs to pick sensors for field robots.\n\nThe soft spot is the lack of numbers. The abstract states the framework generalizes and the results demonstrate robustness, yet it supplies no error rates, standard deviations, or direct comparisons to baselines. Without those, the strength of the no-retuning claim is hard to judge. The key assumption—that keypoint ROI plus SQI stays reliable across posture shifts and distance changes—could be the first thing to fail if the test set missed some edge cases.\n\nThis is for robotics groups or applied sensing teams working on remote monitoring in hazardous environments. A reader who wants implementation ideas and modality limits for robot triage systems will get concrete value.\n\nI would send it to peer review. The real-robot experiments and the platform comparisons are enough to merit referee time even if the quantitative reporting needs tightening.","headline":"This is a straightforward applied systems paper that integrates existing contactless RR methods into a working multimodal setup on three real robot platforms and maps modality ranges, but the abstract gives no accuracy numbers to back the generalization claim.","tokens_in":2326,"tokens_out":442,"would_cite":false,"duration_ms":24140,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.3","headline":"A single multimodal framework lets heterogeneous mobile robots perform contactless respiratory monitoring without platform-specific retuning.","keywords":["contactless respiratory monitoring","mobile robots","multimodal framework","edge computing","respiratory rate","search and rescue","camera modalities","keypoint detection"],"falsifier":"Demonstrating that the respiratory rate estimation fails to maintain accuracy on a fourth robotic platform or under a new combination of lighting and distance without any algorithmic adjustments would falsify the generalization claim.","tokens_in":2703,"feed_emoji":"🤖","tokens_out":634,"duration_ms":38109,"temperature":0.7,"pith_summary":"The paper develops a system that uses multiple camera types on mobile robots to measure breathing rates from a distance. It adapts sensor choice based on brightness and uses keypoint detection to locate the chest area reliably even when subjects change posture. Signal quality filtering ensures accurate estimates. Tests on different robots show the same software works across platforms, with RGB cameras effective up to 8 meters, near-infrared to 6 meters, thermal at short distances, and low-light cameras in darkness up to 8 meters. This approach addresses challenges in emergency situations where direct contact is unsafe.","feed_headline":"Robots monitor breathing contactlessly across platforms up to 8m","feed_subtitle":"One framework adapts RGB, thermal, NIR and low-light cameras on different robots for hazardous environment triage.","key_machinery":"Brightness-adaptive sensor selection across four camera modalities combined with keypoint-guided chest ROI extraction and signal-quality-index filtering for respiratory signal processing.","core_discovery":"The proposed modality-adaptive contactless RR monitoring framework for heterogeneous mobile robots with onboard edge computing generalizes across platforms without per-platform algorithmic retuning. It combines brightness-adaptive sensor selection across RGB, thermal, near-infrared, and low-light cameras with keypoint-guided chest ROI extraction and SQI-based filtering, revealing that RGB covers up to 8m, NIR up to 6m, thermal only at short range, and low-light supports complete darkness up to 8m.","pith_inferences":["The results suggest the system could form the basis for autonomous systems that scan multiple victims in disaster zones.","Modality-specific ranges indicate that hybrid camera setups could optimize coverage in varied environments.","Edge processing on the robot supports real-time decisions during field operations."],"forward_implications":["The framework supports deployment on both quadruped and wheeled robots for remote monitoring.","RGB modality provides the widest operational range up to 8 meters.","Near-infrared remains effective up to 6 meters.","Thermal imaging is limited to short ranges.","Low-light cameras enable monitoring in complete darkness up to 8 meters."],"fun_headline_variants":["Multimodal framework monitors respiration on heterogeneous robots","Contactless breathing monitoring adapts across mobile robot platforms","Robot system combines RGB thermal NIR for respiratory monitoring up to 8m","Edge computing enables contactless RR on varied robots without retuning"],"cache_read_input_tokens":2112,"weakest_assumption_plain":"Keypoint-guided chest ROI extraction combined with SQI-based filtering remains reliable across posture changes, variable illumination, and varying robot-to-subject distances without platform-specific retuning.","fun_headline_variants_meta":{"raw":{"variants":["Multimodal framework monitors respiration on heterogeneous robots","Contactless breathing monitoring adapts across mobile robot platforms","Robot system combines RGB thermal NIR for respiratory monitoring up to 8m","Edge computing enables contactless RR on varied robots without retuning"]},"model":"grok-4.3","cost_usd":0.006985,"raw_usage":{"total_tokens":3272,"prompt_tokens":741,"num_sources_used":0,"completion_tokens":65,"cost_in_usd_ticks":69849500,"prompt_tokens_details":{"text_tokens":741,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":2466,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":741,"tokens_out":65,"duration_ms":29316,"temperature":1.0,"reasoning_tokens":2466,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-06-27T01:36:24.628324+00:00","model_set":{"reader":"grok-4.3"},"falsifier":"Demonstrating that the respiratory rate estimation fails to maintain accuracy on a fourth robotic platform or under a new combination of lighting and distance without any algorithmic adjustments would falsify the generalization claim.","supporting_citations":[],"review_version":1}