Yueyu Hu

About MeI am currently pursuing a Ph.D. with Prof. Yao Wang at NYU Video Lab. I received a master's degree in computer science at Peking University, advised by
                  Prof. Jiaying Liu
                  at STRUCT. My research interests
                  include computer vision, machine learning and image/video compression. I am also interested
                  in
                  photography,
                  trains and railways, meteorology, science-fiction and computer network development.
WHIM
                  as a new part of this site is now online. Check it out!
Friends
                  
Useful LinksWindy:
                    A website to provide visualized numerical weather forecast results by, e.g. GFS, ECMWF.
DLVC:
                      An open-sourced deep-learning driven video codec.
FVC-NIC:
                      Open-Source learned image compression framework.
BDWM:
                      The most popular & official Bulletin Board System (BBS) at Peking University.

                Selected Publications
                
                    Yueyu Hu and Yao Wang.
                    "Learning Neural Volumetric Field for Point Cloud Geometry Compression",
                    2022 Picture Coding Symposium (PCS), 2022. 

                    [Code]
                    [arXiv]
                  
                    Yueyu Hu, Wenhan Yang, Zhan Ma, and Jiaying Liu.
                    "Learning End-to-End Lossy Image Compression: A Benchmark",
                    Accepted by IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2021.

                    [Project]
                    [arXiv]
                    [Early Access]
                    [Code]
                  
                    Yueyu Hu, Wenhan Yang, and Jiaying Liu.
                    "Coarse-to-Fine Hyper-Prior Modeling for Learned Image Compression",
                    Proc. AAAI Conference on Artificial Intelligence (AAAI), 2020.

                    [PDF]
                    [Poster]
                    [Project]
                    [TF Code]
                    [PyTorch Code]
                    [Spotlight]
                  
                    Yueyu Hu, Wenhan Yang, Mading Li, and Jiaying Liu.
                    "Progressive Spatial Recurrent Neural Network for Intra Prediction",
                    IEEE Transactions on Multimedia (TMM), 2019.

                    [PDF]
                    [Code]
                  
                    Yueyu Hu, Shuai Yang, Wenhan Yang, Ling-Yu Duan, and Jiaying Liu.
                    "Towards Coding for Human and Machine Vision: A Scalable Image Coding Approach",
		    Proc. IEEE International Conference on Multimedia & Expo (ICME), 2020. (Best Paper Award)

                    [PDF]
                    [Project]
		    [Slides]
                  
                    Shuai Yang, Yueyu Hu, Wenhan Yang, Ling-Yu Duan, and Jiaying Liu.
                    "Towards Coding for Human and Machine Vision: Scalable Face Image Coding",
                    IEEE Transactions on Multimedia (TMM), 2020.

                    [PDF]
                  
Yueyu Hu, Chunhui Liu, Yanghao Li, Sijie Song and Jiaying Liu. "Temporal
                    Perceptive Network for Skeleton-Based Action Recognition", Proc. British Machine
                      Vision Conference (BMVC), 2017.

                    [PDF]
                  
                    Haofeng Huang, Wenhan Yang, Yueyu Hu, and Jiaying Liu.
                    "Raw-Guided Enhancing Reprocess of Low-Light Image via Deep Exposure Adjustment",
                    Proc. Asian Conference on Computer Vision (ACCV), 2020.
                    
                    [PDF]
                  
                      [Full list of publications]
                  
Projects
                  
                            Towards Coding for Human and Machine Vision: A Scalable Image Coding Approach
                          
                            In this paper, we come up with a novel image coding framework by leveraging both the compressive and the generative models, to support machine vision and human perception tasks jointly. By introducing advanced generative models, we train a flexible network to reconstruct images from compact feature representations and the reference pixels. Experimental results demonstrate the superiority of our framework in both human visual quality and facial landmark detection, which provide useful evidence on the emerging standardization efforts on MPEG VCM (Video Coding for Machine).
                          
                            [PDF]
                            [Supplementary]
                            [Project]
                          
                            Coarse-to-Fine Hyper-Prior Modeling for Learned Image Compression
                          
                           End-to-End learned image compression framework with a coarse-to-fine hyper-prior model,
                            featuring Signal Preserving Hyper Transform and Information Aggregation Reconstruction sub-network.
                            The proposed model achieve superior performance to existing hyper-prior models with context models.
                          
                            [PDF]
                            [Poster]
                            [Project]
                            [Code]
                            [Spotlight]
                          
                            Learning End-to-End Lossy Image Compression: A Benchmark
                          
                            A survey and benchmark of existing end-to-end learned image compression methods.
                          
                            [arXiv]
                            [Project]

                  
Yueyu HU, 胡越予
Ph.D. Student, New York University
 yyhu AT nyu·edu

About MeI am currently pursuing a Ph.D. with Prof. Yao Wang at NYU Video Lab. I received a master's degree in computer science at Peking University, advised by
                  Prof. Jiaying Liu
                  at STRUCT. My research interests
                  include computer vision, machine learning and image/video compression. I am also interested
                  in
                  photography,
                  trains and railways, meteorology, science-fiction and computer network development.
WHIM
                  as a new part of this site is now online. Check it out!

                Selected Publications
                
                    Yueyu Hu and Yao Wang.
                    "Learning Neural Volumetric Field for Point Cloud Geometry Compression",
                    Accepted by 2022 Picture Coding Symposium (PCS), 2022.

                    [Code]
                    [arXiv]
                  
                    Yueyu Hu, Wenhan Yang, Zhan Ma, and Jiaying Liu.
                    "Learning End-to-End Lossy Image Compression: A Benchmark",
                    Accepted by IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2021.

                    [Project]
                    [arXiv]
                    [Early Access]
                    [Code]
                  
                    Yueyu Hu, Wenhan Yang, and Jiaying Liu.
                    "Coarse-to-Fine Hyper-Prior Modeling for Learned Image Compression",
                    Proc. AAAI Conference on Artificial Intelligence (AAAI), 2020.

                    [PDF]
                    [Poster]
                    [Project]
                    [Code]
                    [Spotlight]
        
                    Yueyu Hu, Wenhan Yang, Mading Li, and Jiaying Liu.
                    "Progressive Spatial Recurrent Neural Network for Intra Prediction",
                    IEEE Transactions on Multimedia (TMM), 2019.

                    [PDF]
		    [Code]
                  
                    Yueyu Hu, Shuai Yang, Wenhan Yang, Ling-Yu Duan, and Jiaying Liu.
                    "Towards Coding for Human and Machine Vision: A Scalable Image Coding Approach",
		    Proc. IEEE International Conference on Multimedia & Expo (ICME), 2020. (Best Paper Award)

                    [PDF]
                    [Project]
		    [Slides]
                  
                    Shuai Yang, Yueyu Hu, Wenhan Yang, Ling-Yu Duan, and Jiaying Liu.
                    "Towards Coding for Human and Machine Vision: Scalable Face Image Coding",
                    IEEE Transactions on Multimedia (TMM), 2020.

                    [PDF]
                  
Yueyu Hu, Chunhui Liu, Yanghao Li, Sijie Song and Jiaying Liu. "Temporal
                    Perceptive Network for Skeleton-Based Action Recognition", Proc. British Machine
                      Vision Conference (BMVC), 2017.

                    [PDF]
                  
                    Haofeng Huang, Wenhan Yang, Yueyu Hu, and Jiaying Liu.
                    "Raw-Guided Enhancing Reprocess of Low-Light Image via Deep Exposure Adjustment",
                    Proc. Asian Conference on Computer Vision (ACCV), 2020.
                    
                    [PDF]
                  
                      [Full list of publications]
                  
Projects
                  
                            Towards Coding for Human and Machine Vision: A Scalable Image Coding Approach
                          
                            In this paper, we come up with a novel image coding framework by leveraging both the compressive and the generative models, to support machine vision and human perception tasks jointly. By introducing advanced generative models, we train a flexible network to reconstruct images from compact feature representations and the reference pixels. Experimental results demonstrate the superiority of our framework in both human visual quality and facial landmark detection, which provide useful evidence on the emerging standardization efforts on MPEG VCM (Video Coding for Machine).
                          
                            [PDF]
                            [Supplementary]
                            [Project]
                          
                            Coarse-to-Fine Hyper-Prior Modeling for Learned Image Compression
                          
                            End-to-End learned image compression framework with a coarse-to-fine hyper-prior model,
                            featuring Signal Preserving Hyper Transform and Information Aggregation Reconstruction sub-network.
                            The proposed model achieve superior performance to existing hyper-prior models with context models.
                          
                            [PDF]
                            [Poster]
                            [Project]
                            [TF Code]
                            [PyTorch Code]
                            [Spotlight]
                          
                            Learning End-to-End Lossy Image Compression: A Benchmark
                          
                            A survey and benchmark of existing end-to-end learned image compression methods.
                          
                            [arXiv]
                            [Project]
                          
Friends
                
Useful LinksWindy:
                  A website to provide visualized numerical weather forecast results by, e.g. GFS, ECMWF.
DLVC:
                  An open-sourced deep-learning driven video codec.
FVC-NIC:
                  Open-Source learned image compression framework.
BDWM:
                  The most popular & official Bulletin Board System (BBS) at Peking University.

	Towards Coding for Human and Machine Vision: A Scalable Image Coding Approach In this paper, we come up with a novel image coding framework by leveraging both the compressive and the generative models, to support machine vision and human perception tasks jointly. By introducing advanced generative models, we train a flexible network to reconstruct images from compact feature representations and the reference pixels. Experimental results demonstrate the superiority of our framework in both human visual quality and facial landmark detection, which provide useful evidence on the emerging standardization efforts on MPEG VCM (Video Coding for Machine). [PDF] [Supplementary] [Project]
	Coarse-to-Fine Hyper-Prior Modeling for Learned Image Compression End-to-End learned image compression framework with a coarse-to-fine hyper-prior model, featuring Signal Preserving Hyper Transform and Information Aggregation Reconstruction sub-network. The proposed model achieve superior performance to existing hyper-prior models with context models. [PDF] [Poster] [Project] [Code] [Spotlight]
	Learning End-to-End Lossy Image Compression: A Benchmark A survey and benchmark of existing end-to-end learned image compression methods. [arXiv] [Project]