Unsupervised Domain Adaptation has been an efficient approach to transferring the semantic segmentation model across data distributions. Meanwhile, the recent Open-vocabulary Semantic Scene understanding based on large-scale vision language models is effective in open-set settings because it can learn diverse concepts and categories. However, these prior methods fail to generalize across different camera views due to the lack of cross-view geometric modeling. At present, there are limited studies analyzing cross-view learning. To address this problem, we introduce a novel Unsupervised Cross-view Adaptation Learning approach to modeling the geometric structural change across views in Semantic Scene Understanding. First, we introduce a novel Cross-view Geometric Constraint on Unpaired Data to model structural changes in images and segmentation masks across cameras. Second, we present a new Geodesic Flow-based Correlation Metric to efficiently measure the geometric structural changes across camera views. Third, we introduce a novel view-condition prompting mechanism to enhance the view-information modeling of the open-vocabulary segmentation network in cross-view adaptation learning. The experiments on different cross-view adaptation benchmarks have shown the effectiveness of our approach in cross-view modeling, demonstrating that we achieve State-of-the-Art (SOTA) performance compared to prior unsupervised domain adaptation and open-vocabulary semantic segmentation methods.
Adaptive Multi-view and Temporal Fusing Transformer for 3D Human Pose Estimation
Nanjing University of Information Science and Technology(南京信息工程大学)
本文提出了一种统一的框架称为多视图和时间融合Transformer(MTF-Transformer),以适应不同的视图数和视频长度,而无需摄像机校准的三维人体姿态估计(HPE)。它由特征提取器、多视图融合Transformer(MFT)和时间融合Transformer(TFT)组成。特征提取器从每个图像中估计2D姿态,并根据置信度融合预测。它提供了以姿势为中心的特征嵌入,并使后续模块在计算上轻量化。MFT融合了不同数量的视图与一个新的相对注意力块的功能。它自适应地测量每对视图之间的隐式相对关系,并重建更多的信息特征。TFT聚合整个序列的特征并通过Transformer预测3D姿态。它自适应地处理任意长度的视频,并充分利用时间信息。Transformers的迁移使我们的模型能够更好地学习空间几何,并为不同的应用场景保持鲁棒性。我们报告了Human3.6M、TotalCapture和KTH Multiview Football II的定量和定性结果。与相机参数的最先进的方法相比,MTF变压器获得了有竞争力的结果,并推广到动态捕捉与任意数量的不可见的意见。
Boosting Multi-View Indoor 3D Object Detection via Adaptive 3D Volume Construction
College of Information Science and Electronic Engineering, Zhejiang University(浙江大学信息科学与电子工程学院);Ningbo Global Innovation Center, Zhejiang University(浙江大学宁波全球创新中心);NingboTech University(宁波科技学院);City University of Hong Kong(香港城市大学)
A Geometric-Transformation Feature-Adaptive Manifold Restoration Method for Open-Vocabulary Semantic Segmentation of Remote Sensing Images
School of Remote Sensing and Geomatics Engineering, Nanjing University of Information Science and Technology(南京信息工程大学遥感与测绘工程学院);Nanjing Center, China Geological Survey(中国地质调查局南京中心);School of Automation, Southeast University(东南大学自动化学院)
UniDA3D: A Unified Domain-Adaptive Framework for Multi-View 3D Object Detection
Shenzhen Campus of Sun Yat-sen University(中山大学深圳校区);The State Key Laboratory of Blockchain and Data Security, Zhejiang University(浙江大学区块链与数据安全全国重点实验室);Beijing Academy of Artificial Intelligence(北京人工智能研究院);National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所模式识别国家重点实验室)
EAGLE: Elevating Geometric Reasoning through LLM-empowered Visual Instruction Tuning
Department of Computer Science, Western University(计算机科学系,西部大学);Meituan Inc.(美团公司);Department of Automation, Tsinghua University(自动化系,清华大学);School of Computer Science and Technology, East China Normal University(计算机科学与技术学院,东华大学)
View Adaptive Recurrent Neural Networks for High Performance Human Action Recognition from Skeleton Data
Microsoft Research Asia(微软亚洲研究院);National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所模式识别国家重点实验室)
本文提出了一种基于长短期记忆(LSTM)的自适应聚合递归多视图立体视觉网络,即AA-RMVSNet。我们首先引入了一个视图内聚合模块,通过使用上下文感知卷积和多尺度聚合自适应地提取图像特征,有效地提高了具有挑战性的区域,如薄的物体和大的低纹理表面的性能。为了克服复杂场景中变化遮挡的困难,我们提出了一个视图间成本体积聚合模块,用于自适应像素视图聚合,该模块能够在所有视图之间保持更好的匹配对。提出的两个自适应聚合模块是轻量级的,有效的和互补的提高三维重建的准确性和完整性。代替传统的3D CNN,我们利用具有递归结构的混合网络进行成本体积正则化,这允许高分辨率重建和更精细的假设平面扫描。所提出的网络是端到端训练的,在各种数据集上都取得了优异的性能。该算法在Tanks and Temples基准测试中对$1^{st}$进行了排序,并在DTU数据集上取得了较好的结果,具有较强的泛化能力和鲁棒性。我们的方法的实现可在https://github.com/QT-Zhu/AA-RMVSNet上获得。
Pyramid Multi-view Stereo Net with Self-adaptive View Aggregation
PKU(北京大学);HKU(香港大学);Tencent(腾讯);Kwai Inc.(快手)
本文提出了一种高效的金字塔多视图立体视觉(MVS)网络,该网络具有自适应视图聚合功能,能够实现稠密点云的精确、完整重建。与以往基于深度学习的MVS方法中使用均方方差来生成成本体积不同,我们的\textbf{VA-MVSNet}通过引入两种新的自适应视图聚合:像素视图聚合和体素视图聚合,以较小的额外内存消耗来合并不同视图中的成本方差。为了进一步提高三维点云重建的鲁棒性和完整性,我们扩展了VA-MVSNet与金字塔多尺度图像输入为\textbf{PVA-MVSNet},其中多度量约束被利用来聚合可靠的深度估计在较粗的尺度,以填充在较细的尺度不匹配的区域。实验结果表明,我们的方法建立了一个新的国家的最先进的\textsl{\textbf{DTU}}数据集的完整性和整体质量的显着改善,并通过实现与国家的最先进的方法在\textsl{\textbf{Tanks and Temples}}基准相当的性能具有很强的泛化能力。我们的代码库位于\hyperlink{https://github.com/yhw-yhw/PVAMVSNet}{https://github.com/yhw-yhw/PVAMVSNet}
Joint Adaptive Neighbours and Metric Learning for Multi-view Subspace Clustering
School of Information and Communication Engineering, Dalian University of Technology(大连理工大学信息与通信工程学院);National Laboratory of Pattern Recognition, CASIA(中国科学院自动化研究所模式识别国家重点实验室);University of Chinese Academy of Sciences (UCAS)(中国科学院大学)
Probabilistically Aligned View-unaligned Clustering with Adaptive Template Selection
School of Science, Jiangnan University(江南大学理学院);School of Artificial Intelligence and Computer Science, Jiangnan University(江南大学人工智能与计算机学院);Centre for Vision, Speech and Signal Processing, University of Surrey(萨里大学视觉、语音与信号处理中心)
FrustumFormer: Adaptive Instance-aware Resampling for Multi-view 3D Detection
Institute of Automation, Chinese Academy of Sciences (CASIA)(中国科学院自动化研究所);University of Chinese Academy of Sciences (UCAS)(中国科学院大学);HKISI_CAS(香港创新研究院(中国科学院))
Multi-view Feature Augmentation with Adaptive Class Activation Mapping
School of Computer Science and Technology, University of Chinese Academy of Sciences(中国科学院大学计算机科学与技术学院);Research Center on Fictitious Economy and Data Science, Chinese Academy of Sciences(中国科学院虚拟经济与数据科学研究中心);Key Laboratory of Big Data Mining and Knowledge Management, Chinese Academy of Sciences(中国科学院大数据挖掘与知识管理重点实验室)
Pose Adaptive Dual Mixup for Few-Shot Single-View 3D Reconstruction
Institute of Information Science, Academia Sinica(中央研究院信息科学研究所);Department of Computer Science, National Tsing Hua University(国立清华大学计算机科学系);Department of Computer Science, University of Oxford(牛津大学计算机科学系)
Learning Multi-view Anomaly Detection with Efficient Adaptive Selection
State Key Laboratory of Industrial Control Technology(工业控制技术国家重点实验室);Zhejiang University(浙江大学);Ningbo Innovation Center(宁波创新中心);YouTu Lab(你图实验室);Tencent(腾讯)
Multi-View Adaptive Fusion Network for 3D Object Detection
Jilin University(吉林大学);Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所);University of Chinese Academy of Sciences(中国科学院大学);Sun Yat-Sen University(中山大学);University of Waterloo(滑铁卢大学)
MAGE: View-guided Point Cloud Completion with Efficient Modality Alignment and Adaptive Geometry Enhancement
MAIS, Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所MAIS研究室);School of Artificial Intelligence, University of Chinese Academy of Sciences(中国科学院大学人工智能学院);GIPSA-lab, Univ. Grenoble Alpes, CNRS, Grenoble INP(格勒诺布尔大学GIPSA实验室,法国国家科学研究中心,格勒诺布尔国立综合理工学院)
SeCap: Self-Calibrating and Adaptive Prompts for Cross-view Person Re-Identification in Aerial-Ground Networks
School of Computer Science, Northwestern Polytechnical University(西北工业大学计算机学院);Ningbo Institute, Northwestern Polytechnical University(西北工业大学宁波研究院);Shenzhen Research Institute, Northwestern Polytechnical University(西北工业大学深圳研究院);National Engineering Laboratory for integrated Aero-Space-Ground-Ocean(空天地一体化综合业务网国家工程实验室)
Multi-view Contrastive Learning with Additive Margin for Adaptive Nasopharyngeal Carcinoma Radiotherapy Prediction
The Hong Kong Polytechnic University(香港理工大学);Research Institute for Smart Ageing, The Hong Kong Polytechnic University(香港理工大学智慧老龄研究学院);Nantong University(南通大学)
DA-BEV: Unsupervised Domain Adaptation for Bird's Eye View Perception
Xidian University(西安电子科技大学);Nanyang Technological University(南洋理工大学);University of Technology Sydney(悉尼科技大学);University of Chinese Academy of Sciences(中国科学院大学)