Zhaochen Shi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Zhaochen Shi has a total of 6 co-authors. Here are the top co-authors by number of shared publications:
- David L. Neuhoff
Department of Electrical Engineering and Computer Science, University of Michigan, Ann Arbor, MI 48109
1 shared publication - Huib de Ridder
Perceptual Intelligence Lab, Faculty of Industrial Design Engineering, Delft University of Technology, Delft, the Netherlands
1 shared publication - Jana Zujovic
ECE Department, Northwestern University, Evanston, IL, USA
1 shared publication - Kaixuan Zhang
Department of Electrical and Computer Engineering, Northwestern University, Evanston, IL, USA
1 shared publication - René van Egmond
TU Delft, Delft, Netherlands
1 shared publication - Thrasyvoulos N. Pappas
Northwestern University
1 shared publication
Latest Publications
- 15 February 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 33 (10577149) , 1614-1626
Training and Testing Texture Similarity Metrics for Structurally Lossless Compression
We present a systematic approach for training and testing structural texture similarity metrics (STSIMs) so that they can be used to exploit texture redundancy for structurally lossless image compression. The training and testing is based on a set of image distortions that reflect the characteristics of the perturbations present in natural texture images. We conduct empirical studies to determine the perceived similarity scale across all pairs of original and distorted textures. We then introduce a data-driven approach for training the Mahalanobis formulation of STSIM based on the resulting annotated texture pairs. Experimental results demonstrate that training results in significant improvements in metric performance. We also show that the performance of the trained STSIM metrics is competitive with state of the art metrics based on convolutional neural networks, at substantially lower computational cost.
- 15 February 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 33 (10577149) , 1614-1626
Training and Testing Texture Similarity Metrics for Structurally Lossless Compression
We present a systematic approach for training and testing structural texture similarity metrics (STSIMs) so that they can be used to exploit texture redundancy for structurally lossless image compression. The training and testing is based on a set of image distortions that reflect the characteristics of the perturbations present in natural texture images. We conduct empirical studies to determine the perceived similarity scale across all pairs of original and distorted textures. We then introduce a data-driven approach for training the Mahalanobis formulation of STSIM based on the resulting annotated texture pairs. Experimental results demonstrate that training results in significant improvements in metric performance. We also show that the performance of the trained STSIM metrics is competitive with state of the art metrics based on convolutional neural networks, at substantially lower computational cost.