Joachim Keinert

Fraunhofer Institute for Integrated Circuits

Papers

Total Citations

H-Index

About

Joachim Keinert is a leading researcher in computational imaging and visual computing, with a primary focus on light-field technology, stereo vision, and neural compression. His work bridges the gap between advanced sensor systems and efficient data processing, particularly for immersive applications like virtual reality, autonomous driving, and 3D-TV. Keinert’s major contributions include pioneering the "Non-Planar Inside-Out Dense Light-Field Dataset and Reconstruction Pipeline," a foundational resource that provides precise, dense spatio-angular information for developing and benchmarking reconstruction algorithms. This work is critical for enabling realistic VR experiences and has become a reference in the field. Additionally, his research on "RNNSC: Recurrent Neural Network-Based Stereo Compression Using Image and State Warping" introduces an end-to-end trainable recurrent neural network that efficiently compresses stereo image pairs, addressing the high data demands of multi-camera systems. While his most-cited papers currently hold 4 citations each, their novelty and practical relevance—especially in emerging areas like autonomous navigation and immersive media—signal growing influence. Keinert’s achievements lie in advancing both the theoretical foundations and practical tools for next-generation visual computing, making him a key figure for students and researchers exploring efficient, high-fidelity 3D scene capture and transmission.