Yasushi Sumi

Recognition of 3D free-form objects using segment-based stereo vision

We propose a new method to recognize 3D free-form objects from their apparent contours. It is the extension of our established method to recognize objects with fixed edges. Object models are compared with 3D boundaries which are extracted by segment-based stereo vision. Based on the local shapes of the boundaries, candidate transformations are generated. The candidates are verified and adjusted based on the whole shapes of the boundaries. The models are built from all-around range data of the objects. Experimental results show the effectiveness of the method.

3D Object Recognition Using Segment-Based Stereo Vision

We propose a new method to recognize 3D objects using segment-based stereo vision. Predefined object models are compared with 3D boundaries which are extracted by the stereo vision. Boundaries may be straight lines, circular arcs and free-form curves. The models consist of the local shapes and the whole shapes of the boundaries. The models are constructed from samples of real objects or from CAD models. Based on the local shapes, the candidate transformations are generated. The candidates are verified and adjusted based on the whole shapes. Experimental results show the effectiveness of the method.

Three‐dimensional object recognition using stereo vision

We propose a new method to recognize 3D objects using segment-based stereo vision. This method is available not only for polyhedra but also for objects with free-form boundaries. Predefined object models are compared with 3D boundaries extracted by segment-based stereo vision. The object models consist of local shapes and the whole shapes of the boundaries. The object models are constructed from samples of real objects or from CAD models. Based on the local shapes, candidate transformations are generated. The candidates are verified and adjusted based on the whole shapes. Experimental results show the effectiveness of the method.

A tracking based manipulation system built on stereo vision

This paper presents a manipulation system which is based on visual tracking. A tracking function is useful not only for tracking moving objects but also to verify the result of a task and to detect events in various phases of a task. In our system the tracking function is provided by a stereo vision system. Recognition and tracking procedures are based on a geometric model representation that has generality for a wide range of solid objects. The implemented tracking method is robust against partial occlusion of the tracked object and enables short cycle monitoring of the target coordinates. These features make it possible to build a tracking-based manipulation system. A pick-and-insertion task was executed to show various uses of the tracking function.

Human face analysis based on distributed two-dimensional appearance models

This paper proposes a method for building a system which automatically images the components of a human face. The method is useful in any number of different perspectives. It is assumed that facial appearance depends on the camera angle and perspective as well as individual facial characteristics. Appearances of a face are classified based on the applicable possibilities of the vision algorithms. Two-dimensional models of different facial appearances are constructed and the algorithms based on these models are applied in parallel to each scene in a top-down mode. An appearance model consists of agents, each of which has simple functions. The results of the analysis are re-combined with their assumptions based on the Dempster-Shafer probability model. Experiments using actual facial images confirm the usefulness of the proposed method.