智能音乐生成
简介：通过朱松纯教授AOG理论框架自动生成广告配乐。根据商品的品类、风格等，智能生成与广告视频匹配的背景音乐，以满足商家对广告配乐的个性化可控需求，生成高质量的无版权广告配乐，形成广告视频配乐的商业应用。
研究背景
朱松纯教授团队在计算机视觉领域提出一套完备且强大的人工智能理论框架：AOG（And Or Graph），能够对图像与视频进行时、空、因果的联合解译（Joint Parsing），且可通过MCMC（Markov Chain Monte Carlo）方法采样生成图像/视频。
Joint Parsing: Spatial, Temporal, and Causal Inference for Understanding Images and Videos
图像视频的时空因果联合解译
提出了一个可解释的人工智能（XAI）系统，阐述了一种机器实时理解人类价值观的计算框架，并展示了机器人如何与人类用户通过实时沟通完成一系列复杂人机协作任务，该项研究登上了Science头条。
Realtime Bidirectional human-robot value alignment
实时双向人机价值对齐
提出美学仪表盘（Aesthetic Dashboard），能够对图像视频进行美学评价，包括总分与多项美学属性评分，并且能够推断与经典美学模板的相似程度，以指导拍摄者拍摄出美学质量高的图像视频作品。
Aesthetic Dashboard：Rich Aesthetic Evalution and Guidance for Photography
美学仪表盘：图像视频美学评分与指导
● 论文1
○ 顶刊名称：Science Robitics 2022
○ 论文题目：In situ bidirectional human-robot value alignment
○ 作者：Luyao Yuan, Xiaofeng Gao, Zilong Zheng, Mark Edmonds, Ying Nian Wu, Federico Rossano, Hongjing Lu, Yixin Zhu, Song-Chun Zhu
○ 论文链接：In situ bidirectional human-robot value alignment | Science Robotics
● 论文2
○ 顶会名称：AAAI 2022
○ 论文题目：ValueNet: A New Dataset for Human Value Driven Dialogue System
○ 作者：Liang Qiu, Yizhou Zhao, Jinchao Li, Pan Lu, Baolin Peng, Jianfeng Gao, Song-Chun Zhu
○ 论文链接：ValueNet: A New Dataset for Human Value Driven Dialogue System | Proceedings of the AAAI Conference on Artificial Intelligence
● 论文3
○ 顶会名称：Cognitive Science 2019
○ 论文题目：Partitioning the Perception of Physical and Social Events Within a Unified Psychological Space.
○ 作者：Tianmin Shu, Yujia Peng, Hongjing Lu, Song-Chun Zhu
○ 论文链接：https://cogsci.mindmodeling.org/2019/papers/0194/index.html
● 论文4
○ 顶会名称：ACM Multimedia 2022
○ 论文题目：Attribute Controllable Beautiful Caucasian Face Generation by Aesthetics Driven Reinforcement Learning
○ 作者：Xin Jin, Shu Zhao, Le Zhang, Xin Zhao, Qiang Deng, Chaoen Xiao
○ 论文链接：暂无
● 论文5
○ 顶会名称：ACM Multimedia 2021
○ 论文题目：Aesthetic Evaluation and Guidance for Mobile Photography
○ 作者：Hao Lou, Heng Huang, Chaoen Xiao, Xin Jin
○ 论文链接：Aesthetic Evaluation and Guidance for Mobile Photography | Proceedings of the 29th ACM International Conference on Multimedia
● 论文6
○ 顶刊名称：ACM Transactions on Multimedia Computing Communications and Applications 2022
○ 论文题目：Aesthetic Attribute Assessment of Images Numerically on Mixed Multi-attribute Datasets
○ 作者：Xin Jin, Xinning Li, Hao Lou*, Chenyu Fan, Qiang Deng, Chaoen Xiao*, Shuai Cui, Amit Kumar Singh
○ 论文链接：暂无
● 论文7
○ 顶刊名称：IEEE Transactions on Intelligent Transportation Systems 2022
○ 论文题目：Pseudo-labelling and Meta Reweighting Learning for Image Aesthetic Quality Assessment
○ 作者：Xin Jin, Hao Lou, Heng Huang, Xinning Li, Xiaodong Li, Shuai Cui, Xiaokun Zhang, Xiqiao Li
○ 论文链接：暂无
● 论文8
○ 顶刊名称：IEEE Transactions on Multimedia 2020
○ 论文题目：Deep Multimodality Learning for UAV Video Aesthetic Quality Assessment
○ 作者：Qi Kuang, Xin Jin, Qinping Zhao and Bin Zhou
○ 论文链接：Deep Multimodality Learning for UAV Video Aesthetic Quality Assessment | IEEE Journals & Magazine | IEEE Xplore
