Scene graph generation (SGG) models have suffered from inherent problems regarding the benchmark datasets such as the long-tailed predicate distribution and missing annotation problems. In this work, we aim to alleviate the long-tailed problem of SGG by utilizing unannotated triplets. To this end, we introduce a Self-Training framework for SGG (ST-SGG) that assigns pseudo-labels for unannotated triplets based on which the SGG models are trained. While there has been significant progress in self-training for image recognition, designing a self-training framework for the SGG task is more challenging due to its inherent nature such as the semantic ambiguity and the long-tailed distribution of predicate classes. Hence, we propose a novel pseudo-labeling technique for SGG, called Class-specific Adaptive Thresholding with Momentum (CATM), which is a model-agnostic framework that can be applied to any existing SGG models. Furthermore, we devise a graph structure learner (GSL) that is beneficial when adopting our proposed self-training framework to the state-of-the-art message-passing neural network (MPNN)-based SGG models. Our extensive experiments verify the effectiveness of ST-SGG on various SGG models, particularly in enhancing the performance on fine-grained predicate classes.
RoleMotion: A Large-Scale Dataset towards Robust Scene-Specific Role-Playing Motion Synthesis with Fine-grained Descriptions
University of Science and Technology Beijing（北京科技大学）；Beijing University of Posts and Telecommunications（北京邮电大学）；Shunde Innovation School, University of Science and Technology Beijing（北京科技大学顺德创新学院）；D-Robotics；Linketic；Shandong University（山东大学）；China University of Mining And Technology（中国矿业大学）
Fine-Grained Scene Graph Generation via Sample-Level Bias Prediction
School of Remote Sensing and Information Engineering, Wuhan University（武汉大学遥感信息工程学院）；Ant Group（蚂蚁集团）；College of Computer and Information Technology, China Three Gorges University（三峡大学计算机与信息学院）
Say As You Wish: Fine-grained Control of Image Caption Generation with Abstract Scene Graphs
Renmin University of China（中国人民大学）；Northwestern Polytechnical University（西北工业大学）；Australian Centre for Robotic Vision, University of Adelaide（阿德莱德大学澳大利亚机器人视觉中心）
Fine-Grained Scene Graph Generation with Data Transfer
National University of Singapore（新加坡国立大学）；Tsinghua University（清华大学）；Beijing National Research Center for Information Science and Technology（北京信息科学与技术国家研究中心）
Bipartite Graph Network with Adaptive Message Passing for Unbiased Scene Graph Generation
ShanghaiTech University（上海科技大学）；Shanghai Engineering Research Center of Intelligent Vision and Imaging（上海智能视觉与成像工程技术研究中心）；University of Chinese Academy of Sciences（中国科学院大学）；KU Leuven（鲁汶大学）
CAT-SG: A Large Dynamic Scene Graph Dataset for Fine-Grained Understanding of Cataract Surgery
Chair for Computer-Aided Medical Procedures, Technical University Munich, Germany（计算机辅助医疗程序研究所，慕尼黑技术大学，德国）；Carl Zeiss AG, Munich, Germany（蔡司股份公司，慕尼黑，德国）
3D Gaussian Splatting for Fine-grained Urban Scene Reconstruction via Semantic-guided Density Control and Loss Design
School of Computer Science, China University of Geosciences (Wuhan)（中国地质大学（武汉）计算机学院）；School of Architecture and Urban Planning, Shenzhen University（深圳大学建筑与城市规划学院）；Guangdong Key Laboratory of Urban Informatics, Shenzhen University（深圳大学广东省城市信息学重点实验室）；MNR Key Laboratory for Geo-Environmental Monitoring of Great Bay Area, Shenzhen University（深圳大学自然资源部大湾区地理环境监测重点实验室）
Technical Report for the ICRA 2026 GOOSE 2D Fine-Grained Semantic Segmentation Challenge: Pretraining-Diverse Ensemble of Foundation Vision Encoders for Robust Outdoor Scene Understanding
LION Lab, Hefei University of Technology（合肥工业大学 LION 实验室）；University of Macau（澳门大学）
SceneHGN: Hierarchical Graph Networks for 3D Indoor Scene Generation with Fine-Grained Geometry
Institute of Computing Technology, Chinese Academy of Sciences（中国科学院计算技术研究所）；University of Chinese Academy of Sciences（中国科学院大学）；Stanford University（斯坦福大学）；Cardiff University（卡迪夫大学）
Fine-Grained Scene Image Classification with Modality-Agnostic Adapter
School of Computer Science, Fudan University（复旦大学计算机科学学院）；Innovation Center of Calligraphy and Painting Creation Technology, MCT（中国传媒书画创作技术创新中心）；Videt Technology（维德科技）
Technical Report for ICRA 2026 GOOSE 2D Fine-Grained Semantic Segmentation Challenge: Exploring Query-Based Segmentation and Increased Spatial Context for Outdoor Scene Understanding
Scene-wise Adaptive Network for Dynamic Cold-start Scenes Optimization in CTR Prediction
Huazhong University of Science and Technology（华中科技大学）；School of Software, Beihang University（北京航空航天大学软件学院）；Meituan（美团）；The University of Hong Kong（香港大学）
LSceneLLM: Enhancing Large 3D Scene Understanding Using Adaptive Visual Preferences
South China University of Technology（华南理工大学）；Tencent Robotics X（腾讯Robotics X实验室）；Northeastern University（东北大学）；UMass Amherst（马萨诸塞大学阿默斯特分校）；MIT-IBM Watson AI Lab（麻省理工学院-IBM沃森人工智能实验室）；Pazhou Laboratory（琶洲实验室）；Sichuan University（四川大学）
AGSPNet: A framework for parcel-scale crop fine-grained semantic change detection from UAV high-resolution imagery with agricultural geographic scene constraints
Hunan University of Science and Technology（湖南科技大学）
Uncertainty Estimation for Safety-critical Scene Segmentation via Fine-grained Reward Maximization
The Chinese University of Hong Kong（香港中文大学）；University Hospital of Augsburg（奥格斯堡大学医院）；Harvard Medical School & Massachusetts General Hospital（哈佛医学院与麻省总医院）
Scan2Part: Fine-grained and Hierarchical Part-level Understanding of Real-World 3D Scans
Skolkovo Institute of Science and Technology（斯科尔科沃科学技术学院）
我们提出了Scan 2 Part，一种在现实世界中分割物体各个部分的方法，嘈杂的室内RGB-D扫描。为此，我们在室内场景中的对象的不同的部分层次结构，并探讨其对场景理解模型的影响。具体来说，我们使用一个稀疏的U-Net为基础的架构，通过利用多尺度特征层次结构捕获底层3D扫描几何结构的精细尺度细节。为了训练我们的方法，我们介绍了Scan 2 Part数据集，这是第一个大规模的集合，提供详细的语义标签在现实世界中的设置。在2 cm $^3$和5 cm $^3$两种空间分辨率下，我们总共提供了2，477个ShapeNet对象和1，506个ScanNet场景的53，618个PartNet部分之间的242，081个对应关系。作为输出，我们能够预测细粒度的每个对象部分标签，即使几何体粗糙或部分缺失。
