Shintaro Ono

Fusion of a camera and a laser range sensor for vehicle recognition

This paper presents a system that fuses data from a vision sensor and a laser sensor for detection and classification. Fusion of a vision sensor and a laser range sensor enables us to obtain 3D information of an object together with its textures, offering high reliability and robustness to outdoor conditions. To evaluate the performance of the system, it is applied to recognition of on-street parked vehicles scanned from a moving probe vehicle. The evaluation experiments show obviously successful results, with a detection rate of 100% and an accuracy over 95% in recognizing four vehicle classes.

3D digital archive of the burghers of calais

Auguste Rodin is the most celebrated sculptor of the 19th century. His works, such as The Gates of Hell, The Thinker, and The Burghers of Calais, are famous worldwide. To our knowledge, no art historians have analyzed three-dimensional data of Rodin’s work. This paper describes our project to fill this need by digitally archiving the bronze statue, The Burghers of Calais. First, we scanned the geometrical shape of the sculpture by using a laser range sensor. After that, we analyzed the resulting three-dimensional data using expert knowledge in the field of art history and technology developed in the fields of computer vision and graphics.

Real-time image-based rendering system for virtual city based on image compression technique and eigen texture method

Computer modeling of a large-scale scene such as a city becomes an important topic for computer vision and computer graphics research areas etc. Image-based rendering (IBR) is an effective method for expressing realistic scene, and can construct any arbitrary viewpoint by using the captured real images. However, the large size of the image database in IBR causes serious problems in actual applications, leading to the use of compression techniques. We propose a compression technique based on eigen space combined with a block matching technique to get better result. We also propose a technique to restore the compressed data on Graphic Processing Unit (GPU), allowing us to perform high-speed rendering without raising the load on the CPU.

Space-time Analysis of Spherical Projection Image

In this paper, a novel analysis of space-time volume of spherical projection image is presented. So far, space-time analyses have been extensively conducted for various purposes, i.e. 3-D reconstruction, estimation of camera motion and novel view synthesis and most of them consider only a planer projection and a single camera. In contrast, we conducted analysis on spherical projection for multiple cameras. Since spherical projection does not change its appearance in relation to rotation around the origin of the sphere, extrinsic camera parameters and synchronous parameters of multiple video cameras can be simultaneously estimated by registering multiple space-time volumes of spherical projection, which can be easily achieved by block-matching technique. By using the parameters, multiple video images can be successfully integrated into single omni-directional images without distortions

A probe car for parking-vehicle detection by using laser range sensor

This paper describes a novel method of parking-vehicle detection by using a laser range finder mounted on a probe car. One of the important issues to heavy traffic congestion in Japan is that many vehicles are parked on street and this situation usually decreases the capacity of the street, hence, traffic congestion occurs. Thus it is very important for Japanese road administrators to measure the actual capacity of their road network, including effect of parking vehicles. The method of counting parking vehicles is very simple. A line laser range sensor is vertically mounted on a probe car. As the car runs, we obtain two-dimensional depth image, from which we can reconstruct a set of points in 3D real world. We have developed two kinds of algorithm to count the number of parking vehicles. We have performed outdoor experiments to make sure that our algorithms work good. On the other hand, we have already developed a visualization system for enhanced navigation system: a probe car run in town to obtain various images (textures) of buildings and later we can create a 3-D map of the town with real world texture images. In order to apply this system, we now assume that we have 2-D map of the town. But combining with our new method, we can use 3D range images taken from a probe car instead of given 2-D map. We believe that our measuring method is the first step to town-activity detection.

Classical Sculpture Analysis via Shape Comparison

Modern 3D digital archiving technique provides a reliable assistance to archaeologists. In this paper, we explore a novel framework for digital classical sculpture comparison. Quantitative comparison is introduced, and shape difference is visualized as an indicator to infer unclear points about statue reproduction process in Roman period. Providing a new perspective for the culture relics study, our method plausibly verifies several important assumptions.

Portrait sculptures of Augustus: Categorization via local shape comparison

3D shape comparison with digital copies draws increasing attention in modern culture heritage studies. In this paper, we focus on analyzing portrait sculptures of Augustus with 3D scanned data. A feasible framework of automatic object categorization is proposed based on shape comparison, where distinguishing regions are simultaneously detected as well. High coincidence between our result and previous archaeological speculations is observed in validation experiments, which confirms the validity of the proposed method.

Driving view simulation synthesizing virtual geometry and real images in an experimental mixed-reality traffic space

We propose an efficient and effective image generation system for an experimental mixed-reality traffic space. Our enhanced traffic/driving simulation system represents the view through a hybrid that combines virtual geometry with real images to realize high photo-reality with little human cost. Images for datasets are captured from the real world, and the view for the simulation system is created by synthesizing image datasets - with a conventional driving simulator.

Super resolution of fisheye images captured by on-vehicle camera for visibility support

Fisheye cameras are widely adopted in the back view camera of vehicles and the surveillance camera. Compared to the normal digital cameras, their resolution and image quality are low. In this paper, we propose a method to restore the high resolution images from the fisheye camera by the reconstructed type of super resolution. First, we undistort the fisheye images and regularize these images. Then we super-resolve these regularized images. Blur kernel for super resolution includes the lens blur and the expansion and reduction by the undistortion. The experimental results confirms to the effectiveness of our method which applies to the images captured by the digital camera with fish-eye lens.

On-Vehicle Video Localization Technique based on Video Search using Real Data on the Web

Recently, the mounting of on-vehicle camera is increasing to general cars. Because of this, some users start to upload the on-vehicle videos to web. So that, a number of on-vehicle videos are available nowadays. In this paper, in order to localize car, we propose the efficient matching method for such on-vehicle videos using Temporal Height Image (THI), Affine SIFT and Bag of Feature. THI retains information of relative building height from temporal image sequence. Then we extract robust features from the THI by using Affine SIFT. We realize efficient matching by expressing their features using Bag of features. We conducted experiments to show the efficiency of the proposed method by real image sequences of the city.