Recently, diffusion-based depth estimation methods have drawn widespread attention due to their elegant denoising patterns and promising performance. However, they are typically unreliable under adverse conditions prevalent in real-world scenarios, such as rainy, snowy, etc. In this paper, we propose a novel robust depth estimation method called D4RD, featuring a custom contrastive learning mode tailored for diffusion models to mitigate performance degradation in complex environments. Concretely, we integrate the strength of knowledge distillation into contrastive learning, building the `trinity' contrastive scheme. This scheme utilizes the sampled noise of the forward diffusion process as a natural reference, guiding the predicted noise in diverse scenes toward a more stable and precise optimum. Moreover, we extend noise-level trinity to encompass more generic feature and image levels, establishing a multi-level contrast to distribute the burden of robust perception across the overall network. Before addressing complex scenarios, we enhance the stability of the baseline diffusion model with three straightforward yet effective improvements, which facilitate convergence and remove depth outliers. Extensive experiments demonstrate that D4RD surpasses existing state-of-the-art solutions on synthetic corruption datasets and real-world weather conditions. Source code and data are available at \url{https://github.com/wangjiyuan9/D4RD}.
Digging into Depth Priors for Outdoor Neural Radiance Fields
Tsinghua University(清华大学);Baidu Research(百度研究院);Northwestern Polytechnical University(西北工业大学);Zhejiang University(浙江大学);Institute of Information Engineering, Chinese Academy of Sciences(中国科学院信息工程研究所)
Enhanced Low-resolution LiDAR-Camera Calibration Via Depth Interpolation and Supervised Contrastive Learning
Arizona State University(亚利桑那州立大学);U.S. Army DEVCOM Research Laboratory(美国陆军作战能力发展司令部研究实验室);U.S. Army DEVCOM C5ISR Center(美国陆军作战能力发展司令部指挥、控制、通信、计算机、网络、情报、监视与侦察中心)
Self-supervised Adversarial Training of Monocular Depth Estimation against Physical-World Attacks
Purdue University(普渡大学);University of Missouri – Kansas City(密苏里大学堪萨斯城分校);Rochester Institute of Technology(罗切斯特理工学院);U.S. Naval Research Laboratory(美国海军研究实验室);Meta AI
Attentive and Contrastive Learning for Joint Depth and Motion Field Estimation
Korea Advanced Institute of Science and Technology (KAIST)(韩国科学技术院(KAIST))
从单目视觉系统估计相机的运动以及场景的3D结构是一项复杂的任务,其通常依赖于所谓的场景刚性假设。当观察动态环境时,这一假设被违反,这导致相机的自我运动和对象的运动之间的模糊性。为了解决这个问题,我们提出了一个自监督学习的框架,从单目视频的三维物体运动场估计。我们的贡献是双重的。首先,我们提出了一个两阶段的投影流水线,明确地解开相机自我运动和物体运动与动态注意模块,称为DAM。具体而言,我们设计了一个综合的运动模型,估计在第一和第二翘曲阶段,分别由注意力模块通过共享的运动编码器控制的相机和对象的运动。其次,我们提出了一个对象运动场估计通过对比样本共识,称为CSAC,利用弱语义先验(限界框从对象检测器)和几何约束(每个对象尊重刚体运动模型)。在KITTI,Cityscapes和Waymo Open Dataset上的实验证明了我们方法的相关性,并表明我们的方法在自监督单目深度估计,对象运动分割,单目场景流估计和视觉里程计等任务中优于最先进的算法。
Beihang University(北京航空航天大学);Hangzhou Innovation Institute, Beihang University(北京航空航天大学杭州创新研究院);Institute for Infocomm Research, A*STAR(新加坡科技研究局资讯通信研究院)
CoCoNet: Coupled Contrastive Learning Network with Multi-level Feature Ensemble for Multi-modality Image Fusion
School of Software, Dalian University of Technology(大连理工大学软件学院);Key Laboratory for Ubiquitous Network and Service Software of Liaoning Province(辽宁省普适网络与服务软件重点实验室);Peng Cheng Laboratory(鹏城实验室)
Hybrid-grained Feature Aggregation with Coarse-to-fine Language Guidance for Self-supervised Monocular Depth Estimation
MoE Key Lab of Artificial Intelligence, AI Institute, Shanghai Jiao Tong University(人工智能 MOE 实验室,上海交通大学人工智能学院);Ningbo Institute of Digital Twin, Eastern Institute of Technology, Ningbo, China(宁波数字孪生研究所,东技术研究所,宁波,中国);Ningbo Key Laboratory of Spatial Intelligence and Digital Derivative, Ningbo, China(宁波空间智能与数字衍生关键实验室,宁波,中国);University of Science and Technology of China(中国科学技术大学);CASIA;Tsinghua University(清华大学)
ContrastAlign: Toward Robust BEV Feature Alignment via Contrastive Learning for Multi-Modal 3D Object Detection
School of Computer Science & Technology, Beijing Key Laboratory of Traffic Data Mining and Embodied Intelligence, Beijing Jiaotong University(计算机科学与技术学院,北京交通大数据挖掘与具身智能重点实验室,北京交通大学);Horizon Robotics;Nanyang Technological University(南洋理工大学);University of Macau(澳门大学);School of Information Science and Engineering, Yanshan University(信息科学与工程学院,燕山大学);School of Computer Science and Technology, Harbin Institute of Technology(计算机科学与技术学院,哈尔滨理工大学);School of Information Technology and Electrical Engineering, The University of Queensland(信息技术与电气工程学院,昆士兰大学)
CL-MVSNet: Unsupervised Multi-view Stereo with Dual-level Contrastive Learning
School of Electronic and Computer Engineering, Peking University(北京大学电子与计算机工程学院);Peng Cheng Laboratory(鹏城实验室);Migu Culture Technology Co., Ltd(咪咕文化科技有限公司);School of Computer Science, University of Birmingham(伯明翰大学计算机学院);School of Arts, Peking University(北京大学艺术学院)