Our research team at the Deep Learning and Vision Understanding Group has long been dedicated to advancing visual intelligence across diverse domains. This video presents our ongoing work in Document Image Analysis, where we develop algorithms for optical character recognition, layout parsing, and structured information extraction from scanned documents and digital forms. Next, the video showcases our efforts in Medical Image Processing, focusing on deep learning-based techniques for segmentation, classification, and anomaly detection in radiological images such as CT, MRI, and X-rays, aimed at assisting clinical diagnosis. These two tracks represent our group's sustained commitment to bridging fundamental computer vision research with real-world applications, built on years of collaborative exploration and continuous methodological refinement.

    我们的研究团队隶属于深度学习与视觉理解课题组，长期致力于推动视觉智能在多个领域的应用。视频展示了我们在文档图像分析方面的持续工作——开发用于光学字符识别、版面解析以及从扫描文档和电子表格中提取结构化信息的算法。随后，视频呈现了我们在医学图像处理方面的努力，聚焦于基于深度学习的图像分割、分类和异常检测技术，应用于CT、MRI和X光等放射影像，以辅助临床诊断。这两个方向体现了课题组将基础计算机视觉研究与实际应用相结合的长期坚持，所有成果都源于团队多年来的协作探索与技术积累。