Proceeding of the 5th ACM/IEEE international conference on Human-robot interaction - HRI '10, 2010
ABSTRACT The rehabilitation of the physically challenged is one of the trial applications of the ... more ABSTRACT The rehabilitation of the physically challenged is one of the trial applications of the step-on interface (SOI) on a mobile platform as the friendly amusing mobile (FAM) function. This video shows the result of the preliminary trial.
In the brand-wise random-ordered drinking PET bottles picking task, the overlapping and viewing a... more In the brand-wise random-ordered drinking PET bottles picking task, the overlapping and viewing angle problem makes a low accuracy of the brand recognition. In this paper, we set the problem to increase the brand recognition accuracy and try to find out how the overlapping rate infects on the recognition accuracy. By using a stepping motor and transparent fixture, the training images were taken automatically from the bottles under 360 degrees to simulate a picture taken from viewing angle. After that, the images are augmented with random cropping and rotating to simulate the overlapping and rotation in a real application. By using the automatically constructed dataset, the Inception V3, which was transferred learning from ImageNet, is trained for brand recognition. By generating a random mask with a specific overlapping rate on the original image, the Inception V3 can give 80% accuracy when 45% of the object in the image is visible or 86% accuracy when the overlapping rate is lower than 30%.
ABSTRACT The rehabilitation of the physically challenged is one of the trial applications of the ... more ABSTRACT The rehabilitation of the physically challenged is one of the trial applications of the step-on interface (SOI) on a mobile platform as the friendly amusing mobile (FAM) function. This video shows the result of the preliminary trial.
This paper presents the result of trials of the discrimination of emotion from movement and the a... more This paper presents the result of trials of the discrimination of emotion from movement and the addition of emotion in movement on a teddy bear robot, aiming at both expressing a robot's emotion by movement and improving a robot's personal affinity. We addressed four kinds of emotion - joy, anger, sadness, and fear. In this research, two standpoints were considered
This paper highlights camera orientation estimation accuracy and precision, as well as proposing ... more This paper highlights camera orientation estimation accuracy and precision, as well as proposing a new camera calibration technique using a reinforcement learning method named PPO (Proximal Policy Optimization) in offline mode. The offline mode is used just for extracting the camera geometry parameters that are used for improving accuracy in real-time camera pose estimation techniques. We experiment and compare two popular techniques using 2D vision feedbacks and evaluate their accuracy beside other considerations related to real applications such as disturbance cases from surrounding environment and pose data stability. First, we use feature points detection ORB (Oriented FAST and Rotated BRIEF) and BF (Brute-Force) matcher to detect and match points in different frames, respectively. Second, we use FAST (Features from Accelerated Segment Test) corners and LK (Lucas-Kanade) optical flow methods to detect corners and track their flow in different frames. Those points and corners are then used for the pose estimation through optimization process with the: (a) calibration method of Zhang using chessboard pattern and (b) our proposed method using PPO. The results using our proposed calibration method show significant accuracy improvements and easier deployment for end-user compared to the pre-used methods.
Proceeding of the 5th ACM/IEEE international conference on Human-robot interaction - HRI '10, 2010
ABSTRACT The rehabilitation of the physically challenged is one of the trial applications of the ... more ABSTRACT The rehabilitation of the physically challenged is one of the trial applications of the step-on interface (SOI) on a mobile platform as the friendly amusing mobile (FAM) function. This video shows the result of the preliminary trial.
In the brand-wise random-ordered drinking PET bottles picking task, the overlapping and viewing a... more In the brand-wise random-ordered drinking PET bottles picking task, the overlapping and viewing angle problem makes a low accuracy of the brand recognition. In this paper, we set the problem to increase the brand recognition accuracy and try to find out how the overlapping rate infects on the recognition accuracy. By using a stepping motor and transparent fixture, the training images were taken automatically from the bottles under 360 degrees to simulate a picture taken from viewing angle. After that, the images are augmented with random cropping and rotating to simulate the overlapping and rotation in a real application. By using the automatically constructed dataset, the Inception V3, which was transferred learning from ImageNet, is trained for brand recognition. By generating a random mask with a specific overlapping rate on the original image, the Inception V3 can give 80% accuracy when 45% of the object in the image is visible or 86% accuracy when the overlapping rate is lower than 30%.
ABSTRACT The rehabilitation of the physically challenged is one of the trial applications of the ... more ABSTRACT The rehabilitation of the physically challenged is one of the trial applications of the step-on interface (SOI) on a mobile platform as the friendly amusing mobile (FAM) function. This video shows the result of the preliminary trial.
This paper presents the result of trials of the discrimination of emotion from movement and the a... more This paper presents the result of trials of the discrimination of emotion from movement and the addition of emotion in movement on a teddy bear robot, aiming at both expressing a robot's emotion by movement and improving a robot's personal affinity. We addressed four kinds of emotion - joy, anger, sadness, and fear. In this research, two standpoints were considered
This paper highlights camera orientation estimation accuracy and precision, as well as proposing ... more This paper highlights camera orientation estimation accuracy and precision, as well as proposing a new camera calibration technique using a reinforcement learning method named PPO (Proximal Policy Optimization) in offline mode. The offline mode is used just for extracting the camera geometry parameters that are used for improving accuracy in real-time camera pose estimation techniques. We experiment and compare two popular techniques using 2D vision feedbacks and evaluate their accuracy beside other considerations related to real applications such as disturbance cases from surrounding environment and pose data stability. First, we use feature points detection ORB (Oriented FAST and Rotated BRIEF) and BF (Brute-Force) matcher to detect and match points in different frames, respectively. Second, we use FAST (Features from Accelerated Segment Test) corners and LK (Lucas-Kanade) optical flow methods to detect corners and track their flow in different frames. Those points and corners are then used for the pose estimation through optimization process with the: (a) calibration method of Zhang using chessboard pattern and (b) our proposed method using PPO. The results using our proposed calibration method show significant accuracy improvements and easier deployment for end-user compared to the pre-used methods.
Field and Service Robotics (ISBN: 978-981-15-9460-1), 2021
This paper describes an off-line (i.e. pre-navigation) methodology for machines/robots to identif... more This paper describes an off-line (i.e. pre-navigation) methodology for machines/robots to identify zebra crossings and their respective orientations within pedestrian environments, for the purpose of identifying street crossing ability. Not knowing crossing ability beforehand can prevent path trajectories from being accurately planned pre-navigation. As such, we propose a methodology that sources information from internet 2D maps to identify the locations of pedestrian street crossings. This information is comprised of road networks and satellite imagery of street intersections, from which the locations/orientations of zebra-pattern crossings can be identified by means of trained neural networks and proposed verification algorithms. The methodology demonstrated good capability in detecting and mapping street crossings’ locations, while also showing good results in verifying them against falsely detected objects in satellite imagery. Orientation estimation of zebra-pattern crossings, using a proposed line-scanning algorithm, was found to be within an error range of 4 ∘ on a limited test set.
Uploads
Papers by Takafumi Matsumaru