A curated list of datasets, codes, and papers related to scene change detection (SCD).
NOTE. This repository covers scene change detection based on robot vision (for autonomous driving, drones, mobile robots, etc.). If you are looking for remote sensing change detection (i.e., finding changes between two satellite images), see awesome-remote-sensing-change-detection.
Scene change detection (SCD) refers to the task of localizing changes and identifying change-categories given two scenes, where the two scenes are captured in the same place with a time difference. Often, the scene captured first is called the reference scene (captured at t0), and the scene captured later is called the query scene (captured at t1). A scene can be either an RGB (+D) image, video, or a 3D reconstruction (point cloud). If the scene is an image, SCD is a form of pixel-level prediction because each pixel in the image is classified according to a category. On the other hand, if the scene is point cloud, SCD is a form of point-level prediction because each point in the cloud is classified according to a category.
An example of SCDs from ChangeSim dataset. Detecting scene changes, models are required to distinghish aimed changes (changed objects) from non-targeted changes, i.e., environmental variations (lowered illumination). Changed objects are marked in color.
-
2021 | ChangeSim
IROS -
2020 | Langer et al.
IROS -
2019 | 3RScan
ICCV -
2019 | Mallscape
CVPR -
2017 | Fehr et al.
ICRA
-
2020 | HD map & Localization Dataset (only accessible for Korean researchers due to legal issues)
IROS -
2018 | VL-CMU-CD
Autonomous Robots -
2015 | PCD
BMVC -
2014 | CD2014
CVPRW
- TBU.
- 2021 | DR-TANet: Dynamic Receptive Temporal Attention Network for Street Scene Change Detection
IV - 2020 | Weakly Supervised Silhouette-based Semantic Scene Change Detection
ICRA - 2018 | Learning to measure change: Fully convolutional siamese metric networks for scene change detection
arXiv - 2018 | ChangeNet: A Deep Learning Architecture for Visual Change Detection
ECCVW
- 2021 | Scene change detection: semantic and depth information
MULTIMED TOOLS APPL - 2021 | 3DCD: Scene Independent End-to-End Spatiotemporal Feature Learning Framework for Change Detection in Unseen Videos
T-IP - 2020 | Self-supervised simultaneous alignment and change detection
IROS - 2020 | Better Together: Online Probabilistic Clique Change Detection in 3D Landmark-Based Maps
IROS - 2020 | Hierarchical Paired Channel Fusion Network for Street Scene Change Detection
T-IP - 2020 | Mask-CDNet: A mask based pixel change detection networ
Neurocomputing - 2020 | Scene independency matters: An empirical study of scene dependent and scene independent evaluation for CNN-based change detection
T-ITS - 2020 | CDNet++: Improved Change Detection with Deep Neural Network Feature Correlation
IJCNN - 2019 | Histogram Shape-Based Scene-Change Detection Algorithm
Access
If you have any suggestions about papers, feel free to contact me :)