harpreet

a robot can recognize every object in a room and still not know how to open a single drawer knowing there's a cabinet is scene understanding. knowing which handle to grab, which direction to pull, and that it's a drawer not a door is affordance understanding almost no dataset has it in 3D SceneFun3D: 14,800+ functional elements across 710 laser-scanned indoor scenes. every handle, knob, button, and switch labeled with its affordance, motion axis, motion type, and a natural language task description each scene's 3D point cloud with its iPad video recordings in fiftyone. the affordance annotations live in 3D and project into the video frames check it out here huggingface.co/datasets/Voxe…

boiling changes character before you can see it change. BoilingBench heats a copper surface past the onset of boiling while a high-speed camera, a hydrophone, a microphone, and an acoustic-emission sensor record it. 389,483 frames and 171,970 acoustic hits, synced to surface temperature. i parsed it into FiftyOne so you can scrub the video against the temperature curve. huggingface.co/datasets/Voxe… #PhysicalAI #ComputerVision

deleting several objects one by one leaves a broken scene. CoGeo-GS removes multiple objects from a 3D gaussian scene with geometry-aware editing. the paper says single-object editors often need repeated optimization and leave unstable geometry where objects were removed. arxiv.org/abs/2608.26656 #PhysicalAI #ComputerVision

off pavement, six modalities, labels on the frames. GO Great Outdoors is a large off-road set on a Clearpath Warthog: RGB, stereo, FLIR Boson 640 thermal, Ouster OS1-128 LiDAR, Navtech CTS350-X 2D mmWave radar, NIR, and IMU/GPS. it covers 10.26 km on 5 routes, with 22-class semantic labels on 1800 frames. it is the only set covering all six modalities with labels. CC BY 4.0. arxiv.org/html/2501.19274v3 #PhysicalAI #Robotics

a tactile camera shows the gel deforming. it doesn't tell you how many newtons did it. TaF-Dataset puts the tactile image, a 12x12 pressure array, and a six-axis force sensor on one 30 fps clock. contacts reach 90.8 N. 3,594 episodes, 93 hours. i parsed it into FiftyOne so you can sort episodes by peak force and watch the gel respond. huggingface.co/datasets/Voxe… #PhysicalAI #Robotics

a VR headset as the sensor rig: stereo, depth, and both hands. the Pico annotated set uses a Pico headset and a custom tracker. undistorted stereo video, per-frame depth, dual-channel audio, about 1 kHz 6-DoF head pose, stereo point clouds, and 21-joint hand tracking, with MCAP for Foxglove. coarse actions include approach_shelf, scan_shelf, reach_and_grasp, and inspect_item. 833 GB, CC BY 4.0, gated. scenes include retail and a gymnasium. huggingface.co/datasets/skyc… #PhysicalAI #Robotics

clearance under a bridge is a tight number. DAVE is inland waterway navigation along the Mittellandkanal in Hanover, Germany, for bridge underpass clearance and waterway obstacles. logs are ROS 2 MCAP with multi-beam LiDAR, IP camera images, GNSS, and NMEA wind. zenodo.org/records/13378669 #PhysicalAI #Robotics

Help @Voxel51 make it to the top 10!!

84% of these robot rollouts failed. all 4,000 are in the dataset, labeled. 632 successes and 3,368 failures across 28 manipulation tasks and 99 scenes. every take has a wrist camera, two exterior cameras, and depth. i parsed RoboLab-EgoX into FiftyOne so you can filter to the failures and watch them from every angle. huggingface.co/datasets/Voxe… #PhysicalAI #Robotics

four depth cameras in a ring. the paper asks whether more of them help. the multi-RGBD inertial set uses 4 Intel RealSense D455s, a MicroStrain IMU, and an Ouster-64, with hardware sync under a millisecond. 6 sequences: glare, low light, textureless walls, dynamic occlusion. RA-L 2025, ICRA 2026, with Hyundai Motor Company. reference trajectories are LiDAR-IMU SLAM in TUM format. github.com/seungsang07/multi… #PhysicalAI #Robotics

most robot training runs are wasted on episodes that teach the policy nothing. the fix isn't a bigger model. it's finding the failures that matter and training on those. that's what we built. it's live on product hunt today and we need your help climbing 👇 producthunt.com/products/vox…

a person, a cobot, and a mobile robot can reach the same cell together. the ULTIMATE workshop twin is a synthetic industrial workspace with 3 synced RGB-D streams, a cobot, an AMR, and motion-capture human motion. it has 4 hazard events in 12 ROS 2 Humble bags, 20 GB. zenodo.org/records/17235460 #PhysicalAI #Robotics

falls and daily actions, frame by frame, including people who use wheelchairs. SAFER-Activities is 66 hours of multi-camera monitoring. 85K frame-level action instances across 30 classes, with a wheelchair-user subset. ECCV 2026. tied to healthcare monitoring. on Hugging Face. arxiv.org/abs/2609.08038 #PhysicalAI #MedicalAI

point a lidar at a mirror and it measures a room that isn't there. in an elevator, the cabin appears to extend into solid wall. MirrorSentinel runs a ZED2 camera and an Ouster lidar through 7 cabins. each cabin was measured by hand, so you can check the real boundary against what the sensor reports. i parsed it into FiftyOne so you can scrub the camera and the point cloud together. huggingface.co/datasets/Voxe… #PhysicalAI #Robotics

follow one object in a point cloud without a class name. TETrack3D tracks a single 3D object with temporal key-value caches and a state that evolves over time. github.com/NJU-PCALab/TETrac… #PhysicalAI #ComputerVision

can't be bothered typing slack messages so now i'm gonna just send voice memos

swap the clothes, the person, or the room while the video keeps playing. Vidu-S2 does real-time interactive editable spatial video generation at 720p, with dynamic references and replacement of style, clothing, character, and background. arxiv.org/abs/2609.11638 #PhysicalAI #ComputerVision

similarity metrics trained on photos miss ultrasound speckle. UltraPIPS is a library of perceptual metrics for B-mode ultrasound. standard LPIPS backbones trained on natural images miss speckle and other acoustic statistics. arxiv.org/abs/2608.26033 #PhysicalAI #MedicalAI

the gantry already knows where it is. the radios have to catch up. RoboLoc-G is localization and tracking on an industrial 3-axis CNC gantry with sub-centimeter measured position. it includes 60 GHz TI mmWave radar (IWR6843ISK and IWR6843AOP), Wi-Fi FTM ranging, Pozyx UWB with 9-axis IMUs, a 10-camera OptiTrack system, and 2 external video cameras. ROS 2 bags cover 8 trajectory patterns. zenodo.org/records/15989282 #PhysicalAI #Robotics

most benchmarks in this domain test against synthetic assumptions. real-world deployment breaks on edge cases nobody captures. RGB-P Wild captures the ground truth: 10,000 paired rgb and 4-angle linear polarization sets (0°, 45°, 90°, 135°) for autonomous driving perception under glare, transparency, and specular reflections. arxiv.org/abs/2610.25396v1