Datasets

• Agriculture
  • Strawberry picking point localisation ripeness and weight estimation
  • BACCHUS Long-Term Dataset – Agricultural multi-month dataset with overlapping paths for mapping and localisation algorithms for autonomous robots
  • Long-Term Strawberry Growing Dataset – Images and robot data (GPS, odometry, LiDAR, image streams) collected twice a week over 3 growing seasons
  • Outdoor Action/Intention Recognition Dataset, from the RASberry project
  • Soil aggregate size distribution dataset, for estimating soil structure
  • RGBD datasets of broccoli in the field collected in UK and Spain.
  • RGBD crops versus weeds discrimination dataset, collected in the fields of Lincolnshire
• Medicine
  • Breast Palpation Dataset
  • Needle Insertion Dataset
• Tactile Manipulation
  • Tactile Prediction dataset
  • Slip Control Dataset
  • Video-Tactile prediction dataset
• Indoors
  • L-CAS Office Dataset – visual object detection and recognition, part of the ActiVis project
  • 3D Social Activity Datasets – Datasets for the Recognition of the Human Social Activities using RGB-D and skeleton data.
  • L-CAS 3D Point Cloud Dataset for People Detection, from H2020 project FLOBOT
  • L-CAS Multisensor People Dataset, from H2020 project FLOBOT
  • L-CAS RGB-D-T Re-identification Dataset, from H2020 project ENRICHME
  • L-CAS Thermal Physiological Monitoring Dataset, from H2020 project ENRICHME
  • L-CAS domotic sensors Dataset, from H2020 project ENRICHME
  • L-CAS Datasets for Long-Term Autonomy of Mobile Robots, from FP7 project STRANDS
  • UoL Corridor long-term pedestrian dataset, from H2020 project ILIAD
  • UoL Office long-term pedestrian dataset, from H2020 project ILIAD

Software

Most of our software is open source and free to use for non-commercial purposes.

Members of L-CAS are active contributors to a number of software projects, for instance:

• FreMEn: Frequency-based Map Enhancing
• Contribution to QSR library, mostly around Qualitative Trajectory Calculus
• many contributions to the open source STRANDS repository
• iceWing: A C++ image processing framework tightly integrated with openCV
• altURI: contribution to USARSim
• CAST: The CoSy Architecture Schema Toolkit (CAST) is a system integration framework designed to allow researchers (primarily in the fields of AI and robotics) to develop distributed intelligent, autonomous systems
• MuPPeTS (Multimodal People Perception and Tracking Software): several C++ libraries for multisensor human detection, recognition and tracking. See also our ROS person tracking packages build on top of MuPPETS
• RoboLib: C# software wrappers and practical tutorials for different robots including Rovio, Roomba and NXT.
• EVK1104: Tutorials on code development for Atmel EVK1104 evaluation boards.
• STeF-map: ROS implementation of STeF-map. This is a spatio-temporal flow representation that models the likelihood of people motion direction on a grid-based map by a set of harmonic functions, which efficiently capture long-term variations of crowd movements over time.