Leonid Sigal
45Patents
8h-index
67Co-inventors
74Inventor score
Filing activity: Aug 4, 2004 → Jun 2, 2021
Most-cited inventions
| Patent | Title | Area | Cited by | Status |
|---|---|---|---|---|
| US8786680B2 | Motion capture from body mounted cameras | Physics | 66 | Active |
| US9189886B2 | Method and apparatus for estimating body shape | Physics | 58 | Active |
| US10073861B2 | Story albums | Physics | 32 | Active |
| US7668541B2 | Enhanced techniques for using core based nodes for state transfer | Electricity | 19 | Expired |
| US8823639B2 | Elastomeric input device | Physics | 17 | Active |
| US9058663B2 | Modeling human-human interactions for monocular 3D pose estimation | Physics | 12 | Active |
| US10002460B2 | Method and apparatus for estimating body shape | Physics | 11 | Active |
| US10339706B2 | Method and apparatus for estimating body shape | Physics | 8 | Active |
| US10546417B2 | Method and apparatus for estimating body shape | Physics | 8 | Active |
| US7436980B2 | Graphical object models for detection and tracking | Physics | 7 | Active |
| US9805264B2 | Incremental learning framework for object detection in videos | Physics | 6 | Active |
| US9514363B2 | Eye gaze driven spatio-temporal action localization | Physics | 5 | Active |
| US8886180B2 | Enhanced techniques for using core based nodes for state transfer | Electricity | 4 | Active |
| US10902343B2 | Deep-learning motion priors for full-body performance capture in real-time | Physics | 4 | Active |
| US9940522B2 | Systems and methods for identifying activities and/or events in media contents based on object data and scene data | Physics | 4 | Active |
| US10726206B2 | Visual reference resolution using attention memory for visual dialog | Physics | 4 | Active |
| US9846845B2 | Hierarchical model for human activity recognition | Physics | 3 | Active |
| US9322653B2 | Video-based motion capture and adaptation | Physics | 3 | Active |
| US9477908B2 | Multi-level framework for object detection | Physics | 3 | Active |
| US11409791B2 | Joint heterogeneous language-vision embeddings for video tagging and search | Electricity | 3 | Active |
| US11055537B2 | Systems and methods for determining actions depicted in media contents based on attention weights of media content frames | Physics | 2 | Active |
| US10061986B2 | Systems and methods for identifying activities in media contents based on prediction confidences | Physics | 2 | Active |
| US10558761B2 | Alignment of video and textual sequences for metadata analysis | Physics | 2 | Active |
| US10013621B2 | Systems and methods for identifying objects in media contents | Physics | 1 | Active |
| US9740964B2 | Object classification through semantic mapping | Physics | 1 | Active |
Source: USPTO / EPO open patent data. Inventor disambiguation is heuristic; counts are objective bibliographic measures.