The dataset consists of 45 object images and its corresponding 3D position coordinates. Object images come from the camera of Human Support Robot (HSR) when HSR moves around the virtual environment. Virtual environment is built on GAZEBO and based on the Future Convenience Store Challenge (FCSC), one of the contests for WRS (World Robot Summit). Corresponding 3D position coordinates are from the pose of object model in GAZEBO world file.The virtual environment simulates a supermarket scenario which includes 12 shelves.
The placement of objects is assigned as follows：1fruits, 2drinks, 3cookie, 4can, 5cup noodles, 6curry, 7laundry supplies and cookies, 8tissue, 9tableware,10tools, 11balls, 12one tissue, one cookie, and one drink.
Each directory including the raw image data (.png) and corresponding 3D position coordinates(.csv).
object-raw/ (raw image data)
coordinate_image_csvfile/ (corresponding 3D position coordinates)