No abstract

BackgroundPast studies have shown that robot-based intervention was effective in improving gestural use in children with autism spectrum disorders (ASD). The present study examined whether children with ASD could catch up to the level of gestural production found in age-matched children with typical development and whether they showed an increase in verbal imitation after the completion of robot-based training. We also explored the cognitive and motor skills associated with gestural learning.MethodsChildren with ASD were randomly assigned to two groups. Four- to 6-year-old children with ASD in the intervention group (N = 15) received four 30-min robot-based gestural training sessions. In each session, a social robot, NAO, narrated five stories and gestured (e.g., both hands clapping for an awesome expression). Children with ASD were told to imitate the gestures during training. Age-matched children with ASD in the wait-list control group (N = 15) and age-matched children with typical development (N = 15) received the gestural training after the completion of research. Standardized pretests and posttests (both immediate and delayed) were administered to assess the accuracy and appropriateness of gestural production in both training and novel stories. Children’s language and communication abilities, gestural recognition skills, fine motor proficiencies, and attention skills were also examined.ResultsChildren with ASD in the intervention condition were more likely to produce accurate or appropriate intransitive gestures in training and novel stories than those in the wait-list control. The positive learning outcomes were maintained in the delayed posttests. The level of gestural production accuracy in children with ASD in the delayed posttest of novel stories was comparable to that in children with typical development, suggesting that children with ASD could catch up to the level of gestural production found in children with typical development. Children with ASD in the intervention condition were also more likely to produce verbal markers while gesturing than those in the wait-list control. Gestural recognition skills were found to significantly predict the learning of gestural production accuracy in the children with ASD, with such relation partially mediated via spontaneous imitation.ConclusionsRobot-based intervention may reduce the gestural delay in children with ASD in their early childhood.

This paper proposes a novel method for accurately counting the number of vehicles that are involved in multiplevehicle occlusions, based on the resolvability of each occluded vehicle, as seen in a monocular traffic image sequence. Assuming that the occluded vehicles are segmented from the road background by a previously proposed vehicle segmentation method and that a deformable model is geometrically fitted onto the occluded vehicles, the proposed method first deduces the number of vertices per individual vehicle from the camera configuration. Second, a contour description model is utilized to describe the direction of the contour segments with respect to its vanishing points, from which individual contour description and vehicle count are determined. Third, it assigns a resolvability index to each occluded vehicle based on a resolvability model, from which each occluded vehicle model is resolved and the vehicle dimension is measured. The proposed method has been tested on 267 sets of real-world monocular traffic images containing 3074 vehicles with multiple-vehicle occlusions and is found to be 100% accurate in calculating vehicle count, in comparison with human inspection. By comparing the estimated dimensions of the resolved generalized deformable model of the vehicle with the actual dimensions published by the manufacturers, the root-mean-square error for width, length, and height estimations are found to be 48, 279, and 76 mm, respectively.

While it has been argued that children with autism spectrum disorders are responsive to robot-like toys, very little research has examined the impact of robot-based intervention on gesture use. These children have delayed gestural development. We used a social robot in two phases to teach them to recognize and produce eight pantomime gestures that expressed feelings and needs. Compared to the children in the wait-list control group (N = 6), those in the intervention group (N = 7) were more likely to recognize gestures and to gesture accurately in trained and untrained scenarios. They also generalized the acquired recognition (but not production) skills to human-to-human interaction. The benefits and limitations of robot-based intervention for gestural learning were highlighted. Implications for Rehabilitation Compared to typically-developing children, children with autism spectrum disorders have delayed development of gesture comprehension and production. Robot-based intervention program was developed to teach children with autism spectrum disorders recognition (Phase I) and production (Phase II) of eight pantomime gestures that expressed feelings and needs. Children in the intervention group (but not in the wait-list control group) were able to recognize more gestures in both trained and untrained scenarios and generalize the acquired gestural recognition skills to human-to-human interaction. Similar findings were reported for gestural production except that there was no strong evidence showing children in the intervention group could produce gestures accurately in human-to-human interaction.

This paper presents a novel method for resolving the occlusion of vehicles seen in a sequence of traffic images taken from a single roadside mounted camera. Its concept is built upon a previously proposed vehicle-segmentation method, which is able to extract the vehicle shape out of the background accurately without the effect of shadows and other visual artifacts. Based on the segmented shape and that the shape can be represented by a simple cubical model, we propose a two-step method: first, detect the curvature of the shape contour to generate a data set of the vehicles occluded and, second, decompose it into individual vehicle models using a vanishing point in three dimensions and the set of curvature points of the composite model. The proposed method has been tested on a number of monocular traffic-image sequences and found that it detects the presence of occlusion correctly and resolves most of the occlusion cases involving two vehicles. It only fails when the occlusion was very severe. Further analysis of vehicle dimension also shows that the average estimation accuracy for vehicle width, length, and height are 94.78%, 94.09%, and 95.44%, respectively.