Optical flow is a useful input for various applications, including 3D reconstruction, pose estimation, tracking, and structure-from-motion. Despite its utility, the field of dense long-term tracking, especially over wide baselines, has not been extensively explored. This paper extends the concept of combining multiple optical flows over logarithmically spaced intervals as proposed by MFT. We demonstrate the compatibility of MFT with different optical flow networks, yielding results that surpass their individual performance. Moreover, we present a simple yet effective combination of these networks within the MFT framework. This approach proves to be competitive with more sophisticated, non-causal methods in terms of position prediction accuracy, highlighting the potential of MFT in enhancing long-term tracking applications.
EndoMatcher: Generalizable Endoscopic Image Matcher via Multi-Domain Pre-training for Robot-Assisted Surgery
State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute of Automation, Chinese Academy of Sciences(多模态人工智能系统国家重点实验室,自动化研究所,中国科学院);School of Artificial Intelligence, University of Chinese Academy of Sciences(人工智能学院,中国科学院大学);Department of Pulmonary and Critical Care Medicine, The First Affiliated Hospital of Sun Yat-sen University(中山大学附属第一医院呼吸与危重症医学科);Hong Kong Institute of Science & Innovation(香港创新科技署);Centre of AI and Robotics, Hong Kong Institute of Science & Innovation(人工智能与机器人中心,香港创新科技署)
Deep Learning Reforms Image Matching: A Survey and Outlook
Electronic Information School, Wuhan University(武汉大学电子信息学院);School of Artificial Intelligence and Automation, Huazhong University of Science and Technology(华中科技大学人工智能与自动化学院)
Multi-attention Associate Prediction Network for Visual Tracking
Changchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Science(中国科学院长春光学精密机械与物理研究所);Northwest institute of nuclear technology(西北核技术研究所)
DDN-SLAM: Real-time Dense Dynamic Neural Implicit SLAM
Dalian University of Technology(大连理工大学);Saarland University of Applied Sciences(萨尔应用技术大学);Shanghai Jiao Tong University(上海交通大学);Dalian Maritime University(大连海事大学)
SemGauss-SLAM: Dense Semantic Gaussian Splatting SLAM
School of Automation and Intelligent Sensing, Key Laboratory of System Control and Information Processing, Ministry of Education of China, Shanghai Jiao Tong University, Shanghai(自动化与智能感知学院、系统控制与信息处理重点实验室、中华人民共和国教育部、上海交通大学、上海);University of Cambridge, United Kingdom(剑桥大学、英国)
Active Scout: Multi-Target Tracking Using Neural Radiance Fields in Dense Urban Environments
DEVCOM Army Research Laboratory(DEVCOM陆军研究实验室);Department of Electrical & Systems Engineering and General Robotics, Automation, Sensing and Perception (GRASP) Laboratory at the University of Pennsylvania(电气与系统工程系及通用机器人、自动化、传感和感知(GRASP)实验室,宾夕法尼亚大学)
Advancing Dense Endoscopic Reconstruction with Gaussian Splatting-driven Surface Normal-aware Tracking and Mapping
The Chinese University of Hong Kong (CUHK)(香港中文大学);CUHK Shenzhen Research Institute(香港中文大学深圳研究院);Hong Kong Institute of Science & Innovation, Chinese Academy of Sciences(中国科学院香港创新研究院);Qilu Hospital of Shandong University(山东大学齐鲁医院)
NeSLAM: Neural Implicit Mapping and Self-Supervised Feature Tracking With Depth Completion and Denoising
Institute of Medical Robotics and Department of Automation, Shanghai Jiao Tong University(上海交通大学医疗机器人研究所和自动化系);Key Laboratory of System Control and Information Processing, Ministry of Education, Shang hai 200240, China(教育部系统控制与信息处理重点实验室,上海200240,中国);School of Electrical and Electronic Engineering, Nanyang Technological University(南洋理工大学电子与电气工程学院)
RoMa v2: Harder Better Faster Denser Feature Matching
Linköping University(_linköping大学);Chalmers University of Technology(_chalmers技术大学);University of Amsterdam(_阿姆斯特丹大学);Centre for Mathematical Sciences, Lund University(_数学科学中心,吕勒奥大学)
GateMOT: Q-Gated Attention for Dense Object Tracking
Huazhong University of Science and Technology, Wuhan, China(华中科技大学,武汉,中国);Zhejiang University, Hangzhou, China(浙江大学,杭州,中国);La Trobe University, Melbourne, Australia(拉筹伯大学,墨尔本,澳大利亚)
Dense Scene Multiple Object Tracking with Box-Plane Matching
Tencent Youtu Lab(腾讯优图实验室)
多目标跟踪(MOT)是计算机视觉中的一项重要任务。由于遮挡问题,MOT仍然具有挑战性,特别是在密集场景中。在检测跟踪的框架下,我们提出了盒平面匹配(BPM)方法来提高密集场景下的MOT性能。首先,我们设计了分层聚集判别模型(LADM)来过滤噪声检测。然后,为了正确地关联剩余检测,我们引入全局注意力特征模型(GAFM)来提取外观特征,并使用它来计算历史轨迹和当前检测之间的外观相似性。最后,我们提出了盒平面匹配策略,以实现数据关联,根据运动相似性和外观相似性之间的轨迹和检测。凭借这三个模块的有效性,我们的团队在ACM MM Grand Challenge Hieve 2020中获得了第一名。
QDTrack: Quasi-Dense Similarity Learning for Appearance-Only Multiple Object Tracking
ETH Zurich(苏黎世联邦理工学院);Shanghai AI Laboratory(上海人工智能实验室);Massachusetts Institute of Technology(麻省理工学院);Stanford University(斯坦福大学);UC Berkeley(加州大学伯克利分校)
提出类稠密相似性学习并构建 QDTrack,通过稠密区域对比学习和最近邻关联实现仅外观多目标跟踪,在 BDD100K MOT 上达到新最优且开销极小。
拥挤视频序列中的人体跟踪是视觉场景理解的重要组成部分。不断增加的人群密度对人类的可见性提出了挑战,限制了现有行人跟踪器对更高人群密度的可扩展性。出于这个原因,我们建议使用Crowd of Heads数据集(Crowd of Heads Dataset,简称CrowHD)来振兴头部跟踪,该数据集由9个序列组成,每个序列包含11,463帧,其中有超过2,276,838个头部和5,230个在不同场景中注释的轨道。为了进行评估,我们提出了一个新的度量,IDEucl,来衡量算法的效力,在图像坐标空间中最长的拉伸保持一个独特的身份,从而建立行人人群运动和跟踪算法的性能之间的对应关系。此外,我们还提出了一个新的头部检测器,HeadHunter,这是专为小头部检测在拥挤的场景。我们扩展了HeadHunter与粒子滤波器和基于颜色直方图的头部跟踪重新识别模块。为了将其作为一个强有力的基线,我们将我们的跟踪器与CroHD上现有的最先进的行人跟踪器进行了比较,并证明了其优越性,特别是在身份保护跟踪指标方面。随着重量轻的头部检测器和跟踪器,这是有效的身份保护,我们相信我们的贡献将有助于在密集人群中的行人跟踪的进步。