Institut für Informationsverarbeitung (TNT)

Martin Benjak studied electrical engineering at the Leibniz Universität Hannover with specialization on computer engineering and received his master’s degree in 2019. After graduating he joined Institut für Informationsverarbeitung at Leibniz University Hannover where he currently works as a research assistant towards his PhD.

His current research interests are in the field of machine learning-based video coding and video coding for machines. Within the field of learning-based video coding, he is working on error concealment, spatial scalability and decoder-side optimization. His main focus in video coding for machines is on energy efficient multi-task feature compression.

Martin has served as publication chair for VCIP 2021 and PCS 2024. In June 2024, he worked as a visiting scientist at the National Yang Ming Chiao Tung University in Hsinchu, Taiwan.

Publications

Show recent publications only

Conference Contributions
- Yi-Hsin Chen, Kuan-Wei Ho, Martin Benjak, Jörn Ostermann, Wen-Hsiao Peng
  On the Rate-Distortion-Complexity Trade-offs of Neural Video Coding
  IEEE 26th International Workshop on Multimedia Signal Processing (MMSP), 2024
  BibTeX
- Martin Benjak, Yi-Hsin Chen, Wen-Hsiao Peng, Jörn Ostermann
  Learning-Based Scalable Video Coding with Spatial and Temporal Prediction
  IEEE International Conference on Visual Communications and Image Processing (VCIP), Jeju, Korea, December 2023
  (pdf) BibTeX
- Hong-Sheng Xie, Yi-Hsin Chen, Wen-Hsiao Peng, Martin Benjak, Jörn Ostermann
  Rate Adaptation for Learned Two-layer B-frame Coding without Signaling Motion Information
  IEEE International Conference on Visual Communications and Image Processing (VCIP), Jeju, Korea, December 2023
  BibTeX
- Martin Benjak, Niklas Aust, Yasser Samayoa, Jörn Ostermann
  Neural Network-based Error Concealment for B-Frames in VVC
  2022 IEEE International Symposium on Circuits and Systems (ISCAS), Austin, Texas, USA, 2022
  (pdf) BibTeX
- Martin Benjak, Yasser Samayoa, Jörn Ostermann
  Neural Network Based Error Concealment for VVC
  Proceedings of the 28th IEEE International Conference on Image Processing (ICIP), Anchorage, Alaska, USA, September 2021
  (pdf) BibTeX
- Martin Benjak, Holger Meuel, Thorsten Laude, Jörn Ostermann
  Enhanced Machine Learning-based Inter Coding for VVC
  2021 International Conference on Artificial Intelligence in Information and Communication (ICAIIC) (ICAIIC 2021), April 2021
  (pdf) BibTeX
Journals
- Yi-Hsin Chen, Hong-Sheng Xie, Cheng-Wei Chen, Zong-Lin Gao, Martin Benjak, Wen-Hsiao Peng, Jörn Ostermann
  Maskcrt: Masked conditional residual transformer for learned video compression
  IEEE Transactions on Circuits and Systems for Video Technology, 2024
  BibTeX
Standardisation Contributions
- Martin Benjak, Jörn Ostermann
  Mock-up of MPEG Website Redesign
  17th Meeting of ISO/IEC JTC 1/SC 29/AG 3 Document m69944, November 2024
  BibTeX
- Yi-Hsin Chen, Yi-Chen Yao, Kuan-Wei Ho, Martin Benjak, Wen-Hsiao Peng
  Bitstream Generation and Bit Rate Fitting Results of MaskCRT for CVQM UHD and 4K Sequences
  17th Meeting of ISO/IEC JTC 1/SC 29/AG 5 Document m69870, November 2024
  BibTeX
- Martin Benjak, Jörn Ostermann
  Comparison of VVC and LCEVC with a wide set of configurations for 4K content
  16th Meeting of ISO/IEC JTC 1/SC 29/WG 4 Document m68908, July 2024
  BibTeX
- Yi-Hsin Chen, Chen-Wei Cheng, Martin Benjak, Wen-Hsiao Peng
  Progress report on MaskCRT
  15th Meeting of ISO/IEC JTC 1/SC 29/AG 5 Document m66976, April 2024
  BibTeX
- Martin Benjak, Jörn Ostermann
  Comparison between LCEVC and VVC
  15th Meeting of ISO/IEC JTC 1/SC 29/WG 4 Document m67212, April 2024
  BibTeX
- Yi-Hsin Chen, Cheng-Wei Chen, Zong-Lin Gao, Martin Benjak, Wen-Hsiao Peng
  Results on the Bit Rate Fitting for MaskCRT
  15th Meeting of ISO/IEC JTC 1/SC 29/AG 5 Document m67455, April 2024
  BibTeX
- Yi-Hsin Chen, Zong-Lin Gao, Yi-Chen Yao, Kuan-Wei Ho, Martin Benjak, Wen-Hsiao Peng
  Bitstream Generation and Bit Rate Fitting Results of MaskCRT for CVQM HD Sequences
  15th Meeting of ISO/IEC JTC 1/SC 29/AG 5 Document m68079, April 2024
  BibTeX
- Yi-Hsin Chen, Zong-Lin Gao, Martin Benjak, Wen-Hsiao Peng
  Response to Call for Learning-Based Video Codecs for Study of Quality Assessment by NYCU and LUH
  14th Meeting of ISO/IEC JTC 1/SC 29/AG 5 Document m66163, January 2024
  BibTeX
- Martin Benjak, Jörn Ostermann
  Applications suitable for AI-based data compression
  1st Meeting of ISO/IEC JTC 1/SC 29/WG 2 Document m55424, 2020
  BibTeX