Selection Process

Among the great technical program, two papers will be awarded with the Best Student Paper Award, and the Best Paper Award. The selection process happens in three different stages.

• • 1. First, the Program co-Chairs, select the Top Papers accepted for presentation at the conference. This is done based on the paper’s reviews, meta-reviews and discussion. Thirteen papers were selected this year.

• • 2. Second, the Program co-Chairs, and General co-Chairs select two Best Paper Award Committee co-Chairs, who put together the Committee (see below). Five papers among the thirteen top-rated papers were short-listed to be presented at the Best Paper Session. The remaining eight papers will be highlighted in the Program in the two “Top Paper Sessions”.

• 3. After the Best Paper Session, the two awardees are selected.

Award Candidates

Search-oriented Micro-video Captioning
Liqiang Nie (Harbin Institute of Technology (Shenzhen)), Leigang Qu (Shandong University), Dai Meng (Kuaishou), Min Zhang (Harbin Institute of Technology (Shenzhen)), Qi Tian (Huawei Technologies Ltd.), Alberto del Bimbo (Universita di Firenze)

ConceptBeam: Concept driven target speech extraction
Yasunori Ohishi (NTT Corporation), Marc Delcroix (NTT Corporation), Tsubasa Ochiai (NTT Corporation), Shoko Araki (NTT Corporation), Daiki Takeuchi (NTT Corporation), Daisuke Niizumi (NTT Corporation), Akisato Kimura (NTT Corporation), Noboru Harada (NTT Corporation), Kunio Kashino (NTT Corporation)

DSE-GAN: Dynamic Semantic Evolution Generative Adversarial Network for Text-to-Image Generation
Mengqi Huang (University of Science and Technology of China), Zhendong Mao (University of Science and Technology of China), Penghui Wang (University of Science and Technology of China), Quan Wang (Beijing University of Posts and Telecommunications), Yongdong Zhang (University of Science and Technology of China)

Learnability Enhancement for Low-light Raw Denoising: Where Paired Real Data Meets Noise Modeling
Hansen Feng (Beijing Institute of Technology), Lizhi Wang (Beijing Institute of Technology), Yuzhi Wang (Megvii Technology Inc.), Hua Huang (Beijing Normal University)

Maze: A Cost-Efficient Video Deduplication System at Web-scale
An Qin (Baidu), Mengbai Xiao (Shandong University), Huang Ben (Baidu), Xiaodong Zhang (Ohio State University)

Top Rated Papers

Interactive Video Corpus Moment Retrieval using Reinforcement Learning
Zhixin Ma (Singapore Management University), Chong Wah Ngo (Singapore Management University)

A Multi-view Spectral-Spatial-Temporal Masked Autoencoder for Decoding Emotions with Self-supervised Learning
Rui Li (Shanghai Jiaotong University), Yiting Wang (Cornell University), Wei-Long Zheng (Shanghai Jiao Tong University), Bao-liang Lu (Shanghai Jiaotong University)

TVFormer: Trajectory-guided visual quality assessment on 360º images with Transformers
Li Yang (Beihang University), Mai Xu (Beihang University, Tsinghua University), Tie Liu (Beihang University), Liangyu Huo (Beijing University of Aeronautics and Astronautics), Xinbo Gao (Chongqing University of Post and Telecommunications)

Improving Generalization for Neural Adaptive Video Streaming via Meta Reinforcement Learning
Nuowen Kan (Shanghai Jiaotong University), Yuankun Jiang (Shanghai Jiao Tong University), Chenglin Li (Shanghai Jiao Tong University), Wenrui Dai (Shanghai Jiao Tong University), Junni Zou (Shanghai Jiao Tong University), Hongkai Xiong (Shanghai Jiao Tong University)

MM-ALT: A Multimodal Automatic Lyric Transcription System
Xiangming Gu (National University of Singapore), Longshen Ou (National University of Singapore), Danielle Ong (National University of Singapore), Ye Wang (National University of Singapore)

Semi-supervised Crowd Counting via Density Agency
Hui LIN (Xi’an Jiaotong University), Zhiheng Ma (Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Chinese Academy of Sciences), Xiaopeng Hong (Harbin Institute of Technology), Yaowei Wang (Pengcheng Laboratory), su zhou (Xi’an Jiaotong University)

Im2Oil: Stroke-Based Oil Painting Rendering with Linearly Controllable Fineness Via Adaptive Sampling
Zhengyan Tong (Shanghai Jiaotong University), Xiaohang Wang (Shanghai Jiaotong University), Shengchao Yuan (Shanghai Jiaotong University), Xuanhong Chen (Shanghai Jiao Tong University), Junjie Wang (Shanghai Jiaotong University), Xiangzhong Fang (Shanghai Jiao Tong University)

CACOLIT: Cross-domain Adaptive Co-learning for Imbalanced Image-to-Image Translation
Yijun Wang (Hunan University), Tao Liang (University College London, University of London), Jianxin Lin (Hunan University, Tsinghua University)

Best Paper Award Committee

Elisa Ricci (Unviersity of Trento) – co-Chair
Cees Snoek (University of Amsterdam) – co-Chair
Miriam Redi (Wikimedia Foundation)
Nuno Vasconcelos (University of California San Diego)
Shervin Shirmohammadi (Uuniversity of Ottawa)
Jiaying Liu (Peking University)