Kai Luo

A training-light approach boosts fisheye object detection by 80% by strategically addressing input distortion, feature misalignment, and decision-level fusion with minimal parameter updates.

Robots can now track people and navigate to destinations using a single panoramic model, overcoming the limitations of forward-facing sensors and separate task policies.

Perspective cameras leave robot manipulation policies dangerously blind to occlusions and scene distractors—a vulnerability largely mitigated by routing decoupled panoramic geometry and semantics directly into the action expert.

Angular-to-Cartesian projection distortions have long bottlenecked spherical 3D perception, but explicitly modulating voxel features with spherical range-azimuth geometry outperforms top surround-view models like SurroundOcc and TPVFormer across varied wild environments.

CylindTrack achieves superior identity preservation in panoramic multi-object tracking by effectively modeling depth consistency and leveraging the unique topology of equirectangular images.

Transforming point seeds into temporally consistent object instances, PS-Track sets a new benchmark in multi-object tracking without relying on traditional bounding box annotations.

After decades of uncertainty, the critical temperature of aluminum is pinned down to a narrow 50K range, thanks to a novel simulation framework.

Fisheye cameras can now see the world in 4D, thanks to a new benchmark and method that tackles the unique distortions of spherical projection for improved occupancy tracking.

Achieve 57% lower depth reprojection error and 21% higher cycle inlier ratio for panoramic 3D scene reconstruction by directly modeling Gaussian Opacity Fields on the sphere.