Papers
1
Total Citations
3
H-Index
1
About
Kent Sommer is a researcher whose work sits at the intersection of computer vision and deep learning, with a particular focus on advancing 3D spatial understanding through neural networks. His most notable contribution, "Towards accurate kidnap resolution through deep learning" (2017), tackles the challenging problem of six-degree-of-freedom position regression—essentially teaching a convolutional neural network to determine both location and orientation from visual data. Building on Google's Inception-V4 architecture, Sommer's model achieved a remarkable 22% and 51% relative improvement over prior state-of-the-art position regression CNNs, demonstrating significant advances in robustness for real-world localization tasks. While his citation count of 3 reflects a focused, early-career impact, the technical rigor of this work—particularly its quantitative benchmarks against established methods—marks it as a meaningful contribution to the field of visual localization. Sommer's research is especially relevant for applications in robotics, autonomous navigation, and augmented reality, where accurate spatial awareness is critical. His work exemplifies how targeted architectural innovations can yield substantial performance gains in deep learning-based perception systems.
Research Focus
Key Achievements
Top Papers
- 1Towards accurate kidnap resolution through deep learning3 citations · 2017