Current Vacancies

Postdoctoral Researcher (Research-oriented Role in Surgical Video Analysis & Intelligent Editing)

博士后研究员（前沿视频分析算法研究方向）

岗位职责

1.开展基于多模态大语言模型（如Video-LLaVA、Qwen-VL等）的手术视频分析、智能剪辑与自动摘要的前沿算法研究；

2.针对手术视频特点（长时程、精细操作），探索视频时序定位、事件检测、视频摘要和智能视频编辑的新方法；

3.以第一作者身份在CVPR、ICCV、ECCV、NeurIPS、MICCAI等国际顶级会议或期刊发表高水平论文，申请相关技术专利；

4.熟练运用 HuggingFace Transformers、TimeSformer、VideoMAE 等前沿视频分析框架，快速验证算法创新点；

5.与临床医学专家合作，推动跨学科技术研究成果转化并指导工程团队实现。

任职要求

1.近期获得或即将获得计算机科学、人工智能或相关领域博士学位；

2.在视频分析、视频编辑、视频摘要等领域有高水平论文发表；

3.深入掌握 HuggingFace Transformers、TimeSformer、VideoMAE、SlowFast 等视频分析前沿模型；

4.精通PyTorch等深度学习框架；

5.热爱AI+医疗交叉研究，具备较强的独立科研能力和跨学科沟通能力。

优先考虑条件

1.具备医疗s视频分析尤其是手术视频分析相关项目经验；

2.熟悉自监督学习、对比学习、行为识别、动作定位或视频理解模型；  

3.有大规模模型训练、分布式训练或模型部署优化经验（如TensorRT、ONNX）；  

4.对视频编辑工具与自动化剪辑流程有了解；  

5.具备良好的代码风格与版本控制习惯（Git）。

申请方式

请将个人简历发送至hr02@cair-cas.org.hk。邮件主题请注明应聘[岗位名称]-[姓名]-[官网投递]。 

Job Responsibilities

1.Conduct cutting-edge research in surgical video analysis, intelligent video editing, and automatic summarization using advanced multimodal large language models (Video-LLaVA, Qwen-VL, etc.);

2.Propose novel algorithms specifically tailored for surgical video characteristics (long duration, detailed surgical actions), exploring tasks such as temporal localization, event detection, video summarization, and automated video editing;

3.Publish high-quality papers as the first author in top-tier international conferences or journals (e.g., CVPR, ICCV, ECCV, NeurIPS, MICCAI) and file relevant technology patents;

4.Rapidly prototype and experimentally validate innovative algorithms using leading frameworks like HuggingFace Transformers, TimeSformer, VideoMAE, MMAction2, etc;

5.Collaborate closely with clinical experts to understand clinical needs and guide implementation of interdisciplinary research outcomes.

 Job Requirements

1.Ph.D. degree (recently completed or expected soon) in Computer Science, Artificial Intelligence, or related fields;

2.Strong research track record in video analysis, video summarization, or multimodal video understanding, with first-author publications in top-tier conferences and journals; 

3.Deep theoretical understanding or practical experience with state-of-the-art video analysis models and frameworks, including HuggingFace Transformers, TimeSformer, VideoMAE, SlowFast, UniFormer, etc;

4.Expertise in PyTorch and other deep learning frameworks;

5.Passion for AI-driven healthcare innovation, strong independent research capability, and excellent interdisciplinary communication skills.

Preferred Qualifications

1.Experience or research background in medical imaging analysis, especially surgical video analysis;

2.Familiarity with self-supervised learning, contrastive learning, or generative models (GANs, Diffusion Models); 

3.Experience with large-scale or distributed training, or model deployment optimization (TensorRT, ONNX);

4.Good coding style and familiarity with version control systems (Git).

Application Method

Please send your resume to hr02@cair-cas.org.hk. For the email subject line, please indicate: Application for [Position Name] - [Name] - [Applied via CAIR Official Website].