Hideyuki Kume

Extrinsic Camera Parameter Estimation Using Video Images and GPS Considering GPS Positioning Accuracy

This paper proposes a method for estimating extrinsic camera parameters using video images and position data acquired by GPS. In conventional methods, the accuracy of the estimated camera position largely depends on the accuracy of GPS positioning data because they assume that GPS position error is very small or normally distributed. However, the actual error of GPS positioning easily grows to the 10m level and the distribution of these errors is changed depending on satellite positions and conditions of the environment. In order to achieve more accurate camera positioning in outdoor environments, in this study, we have employed a simple assumption that true GPS position exists within a certain range from the observed GPS position and the size of the range depends on the GPS positioning accuracy. Concretely, the proposed method estimates camera parameters by minimizing an energy function that is defined by using the reprojection error and the penalty term for GPS positioning.

Detection of 3D points on moving objects from point cloud data for 3D modeling of outdoor environments

A 3D modeling technique for an urban environment can be applied to several applications such as landscape simulations, navigational systems, and mixed reality systems. In this field, the target environment is first measured using several types of sensors (laser rangefinders, cameras, GPS sensors, and gyroscopes). A 3D model of the environment is then constructed based on the results of the 3D measurements. In this 3D modeling process, 3D points that exist on moving objects become obstacles or outliers to enable the construction of an accurate 3D model. To solve this problem, we propose a method for detecting 3D points on moving objects from 3D point cloud data based on photometric consistency and knowledge of the road environment. In our method, 3D points on moving objects are detected based on luminance variations obtained by projecting 3D points onto omnidirectional images. After detecting 3D the points based on evaluation value, the points are detected using prior information of the road environment.

Bundle adjustment using aerial images with two-stage geometric verification

A new SfM pipeline that uses aerial images as external references is proposed.Good matches between ground and aerial images are found by two-stage verification.Consistency of orientation and scale from a feature descriptor is locally verified.Outliers are removed by global verification with sampling based bundle adjustment. In this paper, a new pipeline of structure-from-motion for ground-view images is proposed that uses feature points on an aerial image as references for removing accumulative errors. The challenge here is to design a method for discriminating correct matches from unreliable matches between ground-view images and an aerial image. If we depend on only local image features, it is not possible in principle to remove all the incorrect matches, because there frequently exist repetitive and/or similar patterns, such as road signs. In order to overcome this difficulty, we employ geometric consistency-verification of matches using the RANSAC scheme that comprises two stages: (1) sampling-based local verification focusing on the orientation and scale information extracted by a feature descriptor, and (2) global verification using camera poses estimated by the bundle adjustment using sampled matches.

Evaluation of image processing algorithms on vehicle safety system based on free-viewpoint image rendering

Development of algorithms for vehicle safety systems, which support safety driving, takes a long period of time and a huge cost because it requires an evaluation stage where huge combinations of possible driving situations should be evaluated by using videos which are captured beforehand in real environments. In this paper, we address this problem by using free viewpoint images instead of the real images. More concretely, we generate free-viewpoint images from a combination of a 3D point cloud measured by laser scanners and an omni-directional image sequence acquired in a real environment. We basically rely on the 3D point cloud for geometrically correct virtual viewpoint images. In order to remove the holes caused by the unmeasured region of the 3D point cloud and to remove false silhouettes in surface reconstruction, we have developed a technique of free-viewpoint image generation that uses both a 3D point cloud and depth information extracted from images. In the experiments, we have evaluated our framework with a white line detection algorithm and experimental results have shown the applicability of free-viewpoint images for evaluation of algorithms.

Detection of moving objects from point cloud data using photometric consistency and prior knowledge of road environment

3D models of urban environments are constructed based on measured data using several types of sensors such as laser rangefinders, cameras, GPS sensors and gyroscopes. In this 3D modeling process, 3D points on moving objects become obstacles to enable the construction of an accurate 3D model. To solve this problem, this paper has proposed a method for detecting and removing 3D points on moving objects from 3D point cloud data based on photometric consistency and knowledge of the road environment. In our method, 3D points on moving objects are detected based on luminance variations computed by projecting 3D points onto omnidirectional images. After detecting candidates for points on moving objects, points inside a certain size of region determined by prior knowledge of road environment are removed. We show the effectiveness of the method through experiments for a real outdoor environment.

6-DOF Camera Pose Estimation Using Reference Points on an Aerial Image without Altitude Information

A new method for estimating a six-degrees-of-freedom camera pose for a ground-view image using reference points on an aerial image is presented. Unlike typical PnP problems, altitude information is not available for the reference points in our case. The camera pose is estimated by minimizing a cost function defined as the sum of squared distances between observed 2D positions of reference points on a ground-view image and corresponding lines that are projections of 3D vertical lines passing through 2D reference points on an aerial image. The accuracy of the proposed method is evaluated quantitatively in both simulation and real environments. The availability of the proposed method is demonstrated by generating AR images from aerial and ground-view images downloaded from Google Maps and Flickr.