Industrial Technology Research Institute

Hung-Chun Chou

Dong-Chen Tsai

Yu-Ru Huang

Tsann-Tay Tang

Ping-Chang Shih

Chih-Neng Liu

A training data generation device includes a virtual scene generation unit, an orthographic virtual camera, an object-occlusion determination unit, an object-occlusion determination unit and a perspective virtual camera. The virtual scene generation unit is configured for generating a virtual scene, wherein the virtual scene comprises a plurality of objects. The orthographic virtual camera is configured for capturing a vertical projection image of the virtual scene. The object-occlusion determination unit is configured for labeling an occluded-state of each object according to the vertical projection image. The perspective virtual camera is configured for capturing a perspective projection image of the virtual scene. The training data generation unit is configured for generating a training data of the virtual scene according to the perspective projection image and the occluded-state of each object.

Training data generation device, training data generation method using the same and robot arm system using the same

An automatic objects labeling method includes: capturing M consecutive image frames at one station of an assembly line. Performing an object detection step which includes selecting a detection image frame that displays an operation using a work piece against a target object from the M consecutive image frames; and calibrating the position range of the target object in the detection image frame; retracing from the detection image frame to select an Nth retraced image frame from the M consecutive image frames; obtaining a labeled image of the target object from the Nth retraced image frame according to the position range; comparing the labeled image with images of the M consecutive image frames to find at least one other labeled image similar to the target object; and storing both the labeled image and the at least one other labeled image as the same labeled data set.

Objects automatic labeling method and system applying the same

An object pose estimation system, an execution method thereof and a graphic user interface are provided. The execution method of the object pose estimation system includes the following steps. A feature extraction strategy of a pose estimation unit is determined by a feature extraction strategy neural network model according to a scene point cloud. According to the feature extraction strategy, a model feature is extracted from a 3D model of an object and a scene feature is extracted from the scene point cloud by the pose estimation unit. The model feature is compared with the scene feature by the pose estimation unit to obtain an estimated pose of the object.

Growing community of inventors

Taipei, Taiwan

Hung-Chun Chou