Yasutomo Kawanishi

Moving camera background-subtraction for obstacle detection on railway tracks

We propose a method for detecting obstacles by comparing input and reference train frontal view camera images. In the field of obstacle detection, most methods employ a machine learning approach, so they can only detect pre-trained classes, such as pedestrian, bicycle, etc. This means that obstacles of unknown classes cannot be detected. To overcome this problem, we propose a background subtraction method that can be applied to moving cameras. First, the proposed method computes frame-by-frame correspondences between the current and the reference (database) image sequences. Then, obstacles are detected by applying image subtraction to corresponding frames. To confirm the effectiveness of the proposed method, we conducted an experiment using several image sequences captured on an experimental track. Its results showed that the proposed method could detect various obstacles accurately and effectively.

Privacy-Protected Camera for the Sensing Web

We propose a novel concept of a camera which outputs only privacy-protected information; this camera does not output captured images themselves but outputs images where all people are replaced by symbols. Since the people from this output images cannot be identified, the images can be opened to the Internet so that we could observe and utilize the images freely. In this paper, we discuss why the new concept of the camera is needed, and technical issues that are necessary for implementing it.

Estimation of the Attractiveness of Food Photography Focusing on Main Ingredients

This research aims to develop a method to estimate the attractiveness of a food photo. The proposed method extracts two kinds of image features: 1) those focused on the appearance of the main ingredient, and 2) those focused on the impression of the entire food photo. The former is newly introduced in this paper, whereas the latter is based on previous research. The proposed method integrates these image features with a regression scheme to estimate the attractiveness of an arbitrary food photo. We have also built and released a food image dataset composed of images of ten food categories taken from 36 angles named NU FOOD 360x10. The images were assigned target values of their attractiveness through subjective experiments. Experimental results showed the effectiveness of integrating both kinds of image features.

Hand Waving Gesture Detection using a Far-infrared Sensor Array with Thermo-spatial Region of Interest

We propose a method of hand waving gesture detection using a far-infrared sensor array. The far-infrared sensor array captures the spatial distribution of temperature as a thermal image by detecting far-infrared waves emitted from heat sources. The advantage of the sensor is that it can capture human position and movement while protecting the privacy of the target individual. In addition, it works even at night-time without any light source. However, it is difficult to detect a gesture from a thermal image sequence captured by the sensor due to its low-resolution and noise. The problem is that the noise appears as a similar pattern as the gesture. Therefore, we introduce “Spatial Region of Interest (SRoI)” to focus on the region with motion. Also, to suppress the influence of other heat sources, we introduce “Thermal Region of Interest (TRoI)” to focus on the range of the human body temperature. In this paper, we demonstrate the effectiveness of the method through an experiment and discuss its result.

A Study on Estimating the Attractiveness of Food Photography

This paper proposes a method for estimating the attractiveness of food photos in order to assist a user to shoot them attractively. The proposed method extracts both color and shape features from input food images, and then integrates them according to a regression scheme. By this way, the proposed method estimates the attractiveness of an unknown food photo. We also created a food image dataset taken from various 3D-angles for each food category, and set target values of their attractiveness through subjective experiments. Then, we evaluated the performance of the proposed method in two different ways of constructing the attractiveness estimator: One that constructs it for each food category, and the other that constructs a common attractiveness estimator for all food categories. Experimental results showed the effectiveness of the proposed method in addition to the necessity for adaptively selecting the estimator depending on the appearance of foods for further performance improvement.

Pedestrian orientation classification utilizing single-chip coaxial RGB-ToF camera

This paper proposes a method for pedestrian orientation classification. In image recognition, the accuracy is often degraded by the influence of background. In addition, it is also difficult to remove the background and extract only the human body from an image. To overcome these problems, we utilize a single-chip RGB-ToF camera. This camera can acquire RGB and depth images along the same optical axis at the same moment, and thus segmentation of the RGB image becomes easier by using the coaxial depth image. Our proposed method segmented a human body from its background accurately, which lead to the improvement of the accuracy of pedestrian orientation classification.

Tracking Pedestrians Across Multiple Cameras via Partial Relaxation of Spatio-Temporal Constraint and Utilization of Route Cue

We tackle multiple people tracking across multiple non-overlapping surveillance cameras installed in a wide area. Existing methods attempt to track people across cameras by utilizing appearance features and spatio-temporal cues to re-identify people across adjacent cameras. @ However, in relatively wide public areas like a shopping mall, since many people may walk and stay arbitrarily, the spatio-temporal constraint is too strict to reject correct matchings, which results in matching errors. Additionally, appearance features can be severely influenced by illumination conditions and camera viewpoints against people, making it difficult to match tracklets by appearance features. These two issues cause fragmentation of tracking trajectories across cameras. We deal with the former issue by selectively relaxing the spatio-temporal constraint and the latter one by introducing a route cue. We show results on data captured by cameras in a shopping mall, and demonstrate that the accuracy of across-camera tracking can be significantly increased under considered settings.

Estimating the visual variety of concepts by referring to Web popularity

Increasingly sophisticated methods for data processing demand knowledge on the semantic relationship between language and vision. New fields of research like Explainable AI demand to step away from black-boxed approaches and understanding how the underlying semantics of data sets and AI models work. Advancements in Psycholinguistics suggest, that there is a relationship from language perception to how language production and sentence creation work. In this paper, a method to measure the visual variety of concepts is proposed to quantify the semantic gap between vision and language. For this, an image corpus is recomposed using ImageNet and Web data. Web-based metrics for measuring the popularity of sub-concepts are used as a weighting to ensure that the image composition in a dataset is as natural as possible. Using clustering methods, a score describing the visual variety of each concept is determined. A crowd-sourced survey is conducted to create ground-truth values applicable for this research. The evaluations show that the recomposed image corpus largely improves the measured variety compared to previous datasets. The results are promising and give additional knowledge about the relationship of language and vision.

Detection of Similar Geo-Regions Based on Visual Concepts in Social Photos

Travel destination recommendation is useful to support travel. Considering the recommendation of regions within the destination area to visit, it could be difficult for the users to explicitly indicate their preference. Therefore, we considered that it would be more intuitive to recommend regions in the destination area that are similar to a region already well-known to the user. Thus, in this paper, we propose a method for the detection of similar geo-regions based on Visual Concepts in social photos. We report experimental results and analyses by applying the proposed method to the YFCC100M dataset.

Can a Driver Assistance System Determine if a Driver is Perceiving a Pedestrian? - Consideration of the Driver's Visual Adaptation to Illumination Change.

We propose an estimation method of pedestrian detectabilit y considering the driver’s visual adaptation to illumination change. Since it is important for driver assista nce systems to determine if a driver is perceiving a pedestrian or not, estimation of pedestrian detectabilit y y the driver is required. However, previous studies do not consider drastic illumination changes that degra des the detection performance by the driver. We assumed that driver’s visual characteristics change in pro portion to the adaptation period after illumination change. Therefore we constructed estimators correspondin g to different adaptation periods, and estimated the pedestrian detectability by switching them according t o the period. To evaluate the proposed method, we constructed an experimental environment to present a subje ct with illumination changes and conducted an experiment to measure and estimate the pedestrian detectab ility ccording to the adaptation period. Results showed that the proposed method could estimate the pedestri an detectability accurately after the illumination changed drastically.