Mingyi Yang, Luis Herranz, Fei Yang, Luka Murn, Marc Gorriz Blanch, Shuai Wan, et al. (2023). Semantic Preprocessor for Image Compression for Machines. In IEEE International Conference on Acoustics, Speech and Signal Processing.
Abstract: Visual content is being increasingly transmitted and consumed by machines rather than humans to perform automated content analysis tasks. In this paper, we propose an image preprocessor that optimizes the input image for machine consumption prior to encoding by an off-the-shelf codec designed for human consumption. To achieve a better trade-off between the accuracy of the machine analysis task and bitrate, we propose leveraging pre-extracted semantic information to improve the preprocessor’s ability to accurately identify and filter out task-irrelevant information. Furthermore, we propose a two-part loss function to optimize the preprocessor, consisted of a rate-task performance loss and a semantic distillation loss, which helps the reconstructed image obtain more information that contributes to the accuracy of the task. Experiments show that the proposed preprocessor can save up to 48.83% bitrate compared with the method without the preprocessor, and save up to 36.24% bitrate compared to existing preprocessors for machine vision.
|
Saiping Zhang, L. H., Marta Mrak, Marc Gorriz Blanch, Shuai Wan, Fuzheng Yang. (2022). PeQuENet: Perceptual Quality Enhancement of Compressed Video with Adaptation-and Attention-based Network.
Abstract: In this paper we propose a generative adversarial network (GAN) framework to enhance the perceptual quality of compressed videos. Our framework includes attention and adaptation to different quantization parameters (QPs) in a single model. The attention module exploits global receptive fields that can capture and align long-range correlations between consecutive frames, which can be beneficial for enhancing perceptual quality of videos. The frame to be enhanced is fed into the deep network together with its neighboring frames, and in the first stage features at different depths are extracted. Then extracted features are fed into attention blocks to explore global temporal correlations, followed by a series of upsampling and convolution layers. Finally, the resulting features are processed by the QP-conditional adaptation module which leverages the corresponding QP information. In this way, a single model can be used to enhance adaptively to various QPs without requiring multiple models specific for every QP value, while having similar performance. Experimental results demonstrate the superior performance of the proposed PeQuENet compared with the state-of-the-art compressed video quality enhancement algorithms.
|
Petia Radeva, J. Guerrero, & C. Molina. (1998). A Physics-Based Kohonen Ring..
|
C. Molina, G.P. Prause, Petia Radeva, & M. Sonka. (1998). Catheter Path Reconstruction from Biplane Angiography using 3D Snakes..
|
Xose M. Pardo, Petia Radeva, & Juan J. Villanueva. (1999). Self-Training Statistic Snake for Image Segmentation and Tracking..
|
X. Binefa, Petia Radeva, J.A. Cortijo, & J. Garcia. (1998). Contour detection and color influence in defocused environtments..
|
V. Valev, B. Sankur, & Petia Radeva. (1997). Generalized Non-Reducible Descriptors..
|
E Fernandez-Nofrerias, J. Mauri, A. Tovar, L. Cano, E. Martinez, C. Julia, et al. (2001). Correspondencia de las imagenes de angiografia y ecografia intracoronaria: La fusion..
|
V. Valev, & Petia Radeva. (1994). Structural Pattern Recognition by Non-Reducible Descriptors. In Proc. International Workshop on Syntactic and Structural Pattern Recognition..
|
J. Mauri, E Fernandez-Nofrerias, A. Tovar, E. Martinez, L. Cano, V. Valle, et al. (2001). Ecografia Intracoronaria: Un Nou Pas, la Fusio de Imatges amb la Angiografia, el Software. Revista de la Societat Catalana de Cardiologia, XIIIe Congres de la Societat Catalana de Cardiologia, 4(1):48., .
|
V. Valev, & Petia Radeva. (1995). ECG Recognition by Non-Reducible Descriptors..
|
V. Valev, & Petia Radeva. (1995). Constructing Quantitative Non-Reducible Descriptors..
|
David Rotger, Cristina Cañero, Petia Radeva, J. Mauri, E. Fernandez, A. Tovar, et al. (2001). 3D Interactive Visualization and Volumetric Measurements of Coronary Vessels in IVUS..
|
David Rotger, Cristina Cañero, Petia Radeva, J. Mauri, E. Fernandez, A. Tovar, et al. (2001). Advanced Visualization of 3D data of Intravascular Ultrasound Images..
|
Petia Radeva, Cristina Cañero, Juan J. Villanueva, J. Mauri, & E Fernandez-Nofrerias. (2001). 3D Reconstruction of a Stent by Deformable Models..
|