Kimmo Roimela

City-scale landmark identification on mobile devices

With recent advances in mobile computing, the demand for visual localization or landmark identification on mobile devices is gaining interest. We advance the state of the art in this area by fusing two popular representations of street-level image data — facade-aligned and viewpoint-aligned — and show that they contain complementary information that can be exploited to significantly improve the recall rates on the city scale. We also improve feature detection in low contrast parts of the street-level data, and discuss how to incorporate priors on a users position (e.g. given by noisy GPS readings or network cells), which previous approaches often ignore. Finally, and maybe most importantly, we present our results according to a carefully designed, repeatable evaluation scheme and make publicly available a set of 1.7 million images with ground truth labels, geotags, and calibration data, as well as a difficult set of cell phone query images. We provide these resources as a benchmark to facilitate further research in the area.

High dynamic range texture compression

We present a novel compression scheme for high dynamic range textures, targeted for hardware implementation. Our method encodes images at a constant 8 bits per pixel, for a compression ratio of 6:1. We demonstrate that our method achieves good visual fidelity, surpassing DXTC texture compression of RGBE data which is the most practical method on existing graphics hardware. The decoding logic for our method is simple enough to be implemented as part of the texture fetch unit in graphics hardware.

Experience Explorer: A Life-Logging Platform Based on Mobile Context Collection

As mobile smart phones are being equipped with multiple radio interfaces and sensors, they are becoming capable of continuously collecting context information related to the users and the environment. This information can be utilized in different applications and services for providing rich, context-aware experiences. In this paper, we present a client-server platform that enables life logging, via mobile context collection, and processes the data so that meaningful higher-level context can be derived. This can then be mashed-up with 3rd party Internet services, for providing richer social networking and content associations, keeping privacy and security in mind. We present our core use cases, the proposed end-to-end architecture and current implementation, as well as an explorer user interface that visualizes users’ life experience.

Efficient high dynamic range texture compression

We present a novel compression method for high dynamic range (HDR) textures, targeted for future graphics hardware. Identifying that the existing solutions provide either very high image quality or very simple hardware implementation, we aim to achieve both features in a single solution. Our approach improves upon an existing technique by incorporating a simple chrominance coding that allows overall image quality on par with the state of the art, but at a substantially lower encoding and decoding complexity. The end result is what we believe to be an excellent compromise between image quality and efficiency of hardware implementations. We evaluate our compression method using common test images and established HDR image quality metrics. Additionally, we complement these results with error measurements in the CIE L*a*b* color space in order to separately assess the quality of luminance and chrominance information.

The evolution of perspective view in WalkMap

A wearable computer is a potential platform for map applications: it is mobile in nature, and is often equipped with a head-worn display capable of displaying maps of the surrounding area in graphical form. In this paper, we present a map application, called WalkMap, developed for wearable computers. We concentrate on the visual presentation of the map, and propose a visualisation method that is based on the perspective distortion of a regular two-dimensional birds-eye view map. We also describe the results of a field study, and compare the distorted view to a regular two-dimensional view. The results show that while a perspective visualisation is good for some navigational tasks, for some other tasks a regular map is preferred. We then continue by developing the visualisation further, and present a fully adjustable three-dimensional version of the map application, called WalkMap3D. WalkMap3D can display the map area in three dimensions, but is also capable of displaying both traditional two-dimensional and perspective map views.

A Lightweight Platform for Web Mashups in Immersive Mirror Worlds

Cloud City Scene is a lightweight platform that enables visualizations of Web mashups in an immersive mirror-world environment in which annotations blend in with buildings, terrain, and objects, letting users interact with the underlying real-world scene.

Augmented and interactive video playback based on global camera pose

This paper proposes a video playback system that allows user to expend the field of view to surrounding environments that are not visible in the original video frame, arbitrarily change the viewing angles, and see the superimposed point-of-interest (POIs) data in an augmented reality manner during the video playback. The processing consists of two main steps: in the first step, client uploads a video to the GeoVideo Engine, and then the GeoVideo Engine extracts the geo-metadata and returns them back to the client; in the second step, client requests POIs from server, and then the client renders the video with POIs.

A 3D map augmented photo gallery application on mobile device

This paper proposes a 3D map augmented photo gallery mobile application that allows user to virtually transit from 2D image space to the 3D map space, to expand the field of view to surrounding environments that are not visible in the original image, and to change viewing angles among different global registered images during the image browsing. The processing of images consists of two main steps: in the first step, the client application uploads an image to the GeoImage Engine which extracts the geo-metadata and returns them back to the client; in the second step, the client application requests augmented content from server, and then renders 3D view of images on the screen of mobile devices.

Simple Octree Solution for Multi-resolution LiDAR Processing and Visualisation

LiDAR (light detection and ranging) 3D points (both terrestrial and airborne) have recently become a prevalent type of geospatial data. However, this large-scale data poses significant technical challenges for efficient access and visualization. One approach is to generate points with different density levels, i.e. Multi-resolution data sets that can be used for various purposes. This paper presents a simple static tree-based approach and discusses implementation alternatives in favour of speeding up tree building and point sampling processes. This approach is generic and can be applied to other geographic data types as well.

Experience explorer: context-based browsing of personal media

We designed and built a system for browsing digital content and activity data created and gathered with mobile phones. We evaluated the system with 13 users to study the value of the context-based visualizations in real life. In addition to supporting reminiscing, content aggregated on the map revealed life patterns supporting reflection. Aggregation of items from several people also revealed common interests among friends.