
程光亮
客座教授
利物浦大学长聘准教授(Reader Professor), Autonomous Cyber Physical Systems Lab 主任
研究方向
- 图像/视频场景理解
- 机器人-自动驾驶感知算法
- 语言-视觉联合学习
个人简介
现任利物浦大学长聘准教授(Reader Professor), 现任Autonomous Cyber Physical Systems Lab 主任, 2020年获得北京市高级职称。2017年博士毕业于中国科学院自动化研究所模式识别国家重点实验室,2017-2019年于中科院遥感与数字地球研究所进行博士后研究工作,2019-2023担任商汤科技自动驾驶感知团队研发副总监。现主持(PI) 一项Alan Turing Institute 基金 (270万RMB)。目前发表SCI和EI学术论文近70篇,以第一作者或者通讯作者发表论文20余篇,其中包含5篇T-PAMI, 2篇IJCV, 多篇国际顶级期刊论文(TIP, TGRS, T-CSVT, PR, CVIU等)和国际CCF A类会议论文 (CVPR, ICCV, ECCV, MultiMedia等),另有多篇顶级期刊或会议论文在审稿中。美国专利申请和发表数量达20个以上,很多专利已经应用到商汤科技自研的高级辅助驾驶和自动驾驶系统中,并表现出了优异的性能。现主要研究方向为图像/视频场景理解,机器人-自动驾驶感知算法,以及语言-视觉联合学习等。
教育经历
- 中国科学院自动化研究所 硕博连读 模式识别国家重点实验室 研究领域:计算机视觉、模式识别、机器学习、深度学习、遥感图像处理 2012.09-2017.07
- 中国石油大学(华东)信息与控制工程学院 研究领域:计算机视觉、模式识别、机器学习、深度学习、遥感图像处理 专业: 自动化 主要课程:数字电路、模拟电路、自动控制原理、微机原理、C++编程
工作经历
- 英国利物浦大学长聘准教授(Tenure-track Reader) 研究方向:智能驾驶,具身智能(机器人),智能遥感与碳中和等相关方向 2023.01-至今
- 商汤科技开发有限公司 研究总监/部门长(直接主管:石建萍博士) 负责团队规模:20个全职,15个实习生 负责项目: 高级辅助驾驶L2+ 量产项目相机感知研发;L4 自动驾驶红绿灯识别应用研发;本田项目研发 主要研究领域:多任务学习,域迁移研发,模型压缩与量化,语义分割,目标检测跟踪,网络搜索、超大模型能研发等 2019.07-2023.01
- 中国科学院遥感与数字地球研究所 博士后(助理研究员)(指导老师:赵忠明研究员) 主要项目:道路提取项目、变化检测项目、遥感图像飞机检测项目 2017.07-2019.07
研究项目申请经历
- 英国在申项目 国际交流合作项目(与清华大学):“基于视觉的任意类别识别和目标抓取机器人研究”经费:30w RMB 英国EPSRC项目: 具身智能机器人视觉目标抓取与鲁棒性研究,经费:240w RMB(已提交) 英国EPSRC项目: 具身智能机器人视觉目标抓取与鲁棒性研究,经费:240w RMB(已提交)
- 公司级负责项目 ARD项目(本田联合研发项目):复杂道路场景理解+实车演示,项目总收入: 700w RMB ABP项目(本田联合研发项目):车辆-行人轨迹预测+实车演示,项目总收入: 1000w RMB GOD项目(本田联合项目): 路面通用障碍物与交通标志(灯)识别+实车演示,项目总收入:1000w RMB 高级ADAS辅助驾驶量产,感知团队负责人,负责广汽、合众、一汽项目研发,团队成员 40人 自动驾驶感知算法国家标准制定小组成员(公司研发代表)
- 博士-博士后负责与参与项目 参与了“浦江一号卫星”图像处理模块的算法设计和编程实现,子模块负责人。主要职责:负责快速云判 检测、海陆分离以及舰船检测算法及具体实现。该卫星能够实时在轨处理,是国内首个快响卫星在轨图 像处理平台。 参与了项目组“星簇项目的算法验证”,算法主要负责人,提出了一种全卷积的飞机检测算法,该算法运 行速度为普通算法的35倍,参数仅为普通算法的1/14,85%召回率条件下的误检率为4%,在数据集上 得到了最优性能。 参与了“国家自然科学基金重大项目”,主要算法负责人。主要职责:负责高光谱图像分类算法研究。提 出了一种基于判别分析和鲁棒回归的高光谱图像分类算法。在极少样本点的情况下,该方法在4 个公开 数据集上均取得了最优性能。 参与了“国家自然科学基金项目”,项目实际负责人。主要职责:负责高分辨率遥感图像道路分割与中心 线提取算法研究。提出了一种基于端到端地级联反卷积网络,在公开数据集上取得了最优性能。
期刊编辑&审稿经历
- 期刊编辑:Neurocomputing 期刊编委(中科院二区),Remote Sensing 期刊编辑(中科院二区), Sustainability 期刊编辑(JCR Q2), ICCV2023 领域主席 + ICIP2024 领域主席, Valse 组委会成员
- 期刊审稿: IEEE TPAMI, IEEE TIP, IEEE TGRS, IEEE TITS, ISPRS, Neurocomputing, GRSL, TNNLS etc.
- 会议审稿: CVPR (2020, 2021, 2022, 2023), ICCV(2021, 2023), ECCV (2020, 2022), AAAI (2021,2022), WACV (2020, 2021, 2022), BMVC (2021, 2022, 2023)
博士、硕士培养
- 联合培养 4 名博士(1 名北大,2 名上交,1 名北航),6 名硕士(4 名北航,1 名天大,1 名清华,)
学术竞赛获奖
- 带领团队在国际视频目标分割任务中获得第三名(近100个国际顶级团队参与)
- 获得国内人工智能大赛-车道线检测赛道算法组第一名(50+个国内外团队参赛)
- 曾获得计算机视觉公开数据集PASCAL VOC分割任务第一名(2016.10)
- 自动驾驶公开数据集CityScapes分割任务榜单第一名(2019.8)
获奖经历
- 北京市副教授职称 2020
- 北京市优秀毕业生 (5/140) 2017
- 中科院三好学生 (10/140) 2016
- 攀登奖学金一等奖 (12/700) 2016
- 博士国奖奖学金候选人 (20/700) 2015
- 山东省优秀毕业生(2/130) 2012
- 山东省优秀本科毕业论文(2/130) 2012
- 连续两年获得国家奖学金 (1/130) 2009-2011
项目经历
- 高级辅助驾驶 L2+量产 L2+项目 & 自动驾驶 L4 红绿灯感知应用项目 2019.07~至今
- 商汤-本田自动驾驶联合研究项目(项目负责人) 2019.7~2022.10
教育经历
- Transformer-based visual segmentation: A survey IEEE Transactions on Pattern Analysis and Machine Intelligence (Under Review) 2023
- Change Detection Methods for Remote Sensing in the Last Decade: A Comprehensive Review IEEE Geoscience and Remote Sensing Magazine (Under Review) 2023
- Learn by Oneself: Exploiting Weight-Sharing Potential in Knowledge Distillation Guided Ensemble Network IEEE Transactions on Circuits and Systems for Video Technology 2023
- Tube-link: A flexible cross tube baseline for universal video segmentation IEEE International Conference on Computer Vision (Under Review) 2023
- Local-to-Global Information Communication for Real-Time Semantic Segmentation Network Search IEEE Transactions on Image Processing (Under Review) 2023
- Self-adversarial disentangling for specific domain adaptation IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023 2023
- PanopticPartFormer++: A Unified and Decoupled View for Panoptic Part Segmentation IEEE Transactions on Pattern Analysis and Machine Intelligence (Under Major Revision) 2023
- Betrayed by Captions: Joint Caption Grounding and Generation for Open Vocabulary Instance Segmentation IEEE International Conference on Computer Vision (Under Review), 2023 2023
- Panoptic-PartFormer: Learning a unified model for Panoptic Part Segmentation European Conference on Computer Vision (ECCV) 2022
- Fashionformer: A simple, Effective and Unified Baseline for Human Fashion Segmentation and Recognition European Conference on Computer Vision (ECCV) 2022
- PolyphonicFormer: Unified Query Learning for Depth-aware Video Panoptic Segmentation European Conference on Computer Vision (ECCV) 2022
- Query Learning of Both Thing and Stuff for Panoptic Segmentation IEEE International Conference on Image Processing 2022
- Dmt: Dynamic mutual training for semi-supervised learning Pattern Recognition 2022
- Context-aware mixup for domain adaptive semantic segmentation IEEE Transactions on Circuits and Systems for Video Technology 2022
- Uncertainty-aware consistency regularization for cross-domain semantic segmentation Computer Vision and Image Understanding 2022
- SFNet: Faster, Accurate, and Domain Agnostic Semantic Segmentation via Semantic Flow International Journal of Computer Vision 2022
- TransVOD: End-to-end Video Object Detection with Spatial-Temporal Transformers IEEE Transactions on Pattern Analysis and Machine Intelligence 2022
- Spatio-Temporal Fusion-based Monocular 3D Lane Detection British Machine Vision Conference 2022
- Multi-level Domain Adaptation for Lane Detection IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshop 2022
- Video K-Net: A Simple, Strong, and Unified Baseline for Video Segmentation IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2022
- End-to-End Video Object Detection with Spatial-Temporal Transformers/span> ACM Multimedia 2021
- Global aggregation then local distribution for scene parsing IEEE Transactions on Image Processing 2021
- Towards efficient scene understanding via squeeze reasoning IEEE Transactions on Image Processing 2021
- Improving Video Instance Segmentation via Temporal Pyramid Routing IEEE Transactions on Pattern Analysis and Machine Intelligence 2021
- Boundarysqueeze: Image segmentation as boundary squeezing International Journal of Computer Vision 2021
- Embedded Knowledge Distillation in Depth-Level Dynamic Neural Network IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshop 2021
- Pit: Position-invariant transform for cross-fov domain adaptation IEEE International Conference on Computer Vision (ICCV) 2021
- Enhanced boundary learning for glass-like object segmentation IEEE International Conference on Computer Vision (ICCV) 2021
- Pointflow: Flowing semantics through points for aerial image segmentation IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2021
- Semi-supervised semantic segmentation via dynamic self-training and class-balanced curriculum Arxiv 2020
- Improving Semantic Segmentation via Decoupled Body and Edge Supervision European Conference on Computer Vision (ECCV) 2020
- Search what you want: Barrier penalty NAS for mixed precision quantization European Conference on Computer Vision (ECCV) 2020
- Low-bit quantization needs good distribution IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshop 2020
- Graph-guided architecture search for real-time semantic segmentation IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2020
- Deep similarity fusion networks for one-shot semantic segmentation Asian Conference on Pattern Recognition 2019
- Recognizing road from satellite images by structured neural network Elsevier Neurocomputing 2019
- OVSNet: Towards One-Pass Real-Time Video Object Segmentation ArXiv 2019
- Distinguishing cloud and snow in satellite images via deep convolutional network IEEE Geoscience and Remote Sensing Letters 2017
- Automatic Road Segmentation and Centerline Extraction via Cascaded End-to-end Deconvolution Neural Network IEEE Transactions on Geoscience and Remote Sensing 2017
- Accurate Urban Road Centerline Extraction from VHR Imagery via Multiscale Segmentation and Tensor Voting Elsevier Neurocomputing 2016
- Road Centerline Extraction via Semisupervised Segmentation and Multidirection Nonmaximum Suppression IEEE Geoscience and Remote Sensing Letters 2016
- Semantic segmentation with modified deep residual networks Chinese Conference on Pattern Recognition (CCPR) 2016
- Building extraction from multi-source remote sensing images via deep deconvolution neural networks IEEE International Geoscience and Remote Sensing Symposium (IGRSS) 2016
- Fast aircraft detection using end-to-end fully convolutional network IEEE International Conference on Digital Signal Processing (DSP) 2016
- Semi-supervised Hyperspectral Image Classification via Discriminant Analysis and Robust Regression IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing (JSTARS) 2015
- Road extraction via adaptive graph cuts with multiple features IEEE International Conference on Image Processing (ICIP) 2015
- Urban road extraction via graph cuts based probability propagation IEEE International Conference on Image Processing (ICIP) 2014