彭军

基本信息

彭军，人工智能学院助理教授。

邮箱：pengjun@xmu.edu.cn

研究方向: 高效多模态学习；视觉问答；文生图；视频理解等

履历

2026-至今：厦门大学，人工智能学院，助理教授

2023-2026：鹏城国家实验室，博士后

2017-2023：厦门大学，计算机科学与技术专业，博士学位

2013-2017：厦门大学，计算机科学与技术专业，学士学位

论文清单

1. QCA: Query-and content-aware keyframe selection for long video understanding.

   Jun Peng, Baiyang Song, Jie Li, Hui Li, Yiyi Zhou, Rongrong Ji, Yonghong Tian. ECCV 2026

2. KTV: Keyframes and key tokens selection for efficient training-free video LLMs.

   Baiyang Song, Jun Peng*, Yuxin Zhang, Guangyao Chen, Feidiao Yang, Jianyuan Guo. AAAI 2026

3. Textrefiner: Internal visual feature as efficient refiner for vision-language models prompt tuning.

   Jingjing Xie, Yuxin Zhang, Jun Peng, Zhaohong Huang, Liujuan Cao. AAAI 2025

4. Pixelface+: Towards controllable face generation and manipulation with text descriptions and segmentation masks.

   Xiaoxiong Du, Jun Peng, Yiyi Zhou, Jinlu Zhang, Siting Chen, Guannan Jiang, Xiaoshuai Sun, Rongrong Ji. ACM MM 2023

5. Learning dynamic prior knowledge for text-to-face pixel synthesis.

   Jun Peng, Xiaoxiong Du, Yiyi Zhou, Jing He, Yunhang Shen, Xiaoshuai Sun, Rongrong Ji. ACM MM 2022

6. Towards open-ended text-to-face generation, combination and manipulation.

   Jun Peng, Han Pan, Yiyi Zhou, Jing He, Xiaoshuai Sun, Yan Wang, Yongjian Wu, Rongrong Ji. ACM MM 2022

7. Knowledge-driven generative adversarial network for text-to-image synthesis.

   Jun Peng, Yiyi Zhou, Xiaoshuai Sun, Liujuan Cao, Yongjian Wu, Feiyue Huang, Rongrong Ji. IEEE TMM 2021