Jiaxin Cai

Papers

Total Citations

H-Index

About

Jiaxin Cai is a researcher advancing the frontier of efficient computer vision, with a focus on enabling intelligent systems to operate under severe computational constraints. His primary research areas include video semantic segmentation, compressed video analysis, and resource-aware deep learning for robotics. Cai’s most notable contribution, "Efficient Semantic Segmentation for Compressed Video" (2024), addresses a critical bottleneck in real-world robotics: the need to process high-resolution, high-bit-rate video streams that must be compressed before analysis due to limited onboard computing power. He proposes a novel segmentation paradigm that works directly on compressed video, bypassing the costly decompression step and dramatically reducing processing latency. This work has already garnered early citations, signaling its importance to the field. By tackling the intersection of video compression and semantic understanding, Cai is helping to make autonomous systems—from drones to mobile robots—more responsive and energy-efficient. His research holds promise for practical deployments where every millisecond and watt matters, positioning him as an emerging voice in efficient vision for embedded AI.