Four Eyes Lab Open House

March 15, 2006, 2-5pm

Location: Four Eyes Lab, Trailer 935

(Directly east of Phelps Hall, the middle of three trailers)

http://ilab.cs.ucsb.edu

Click on map
for location

Mobile Augmented Reality - Towards "Anywhere Augmentation"

Jason Wither

The goal of this project is to allow users working in augmented reality to interact with and augment a real world outdoor scene that has not already been modeled. We strive to accomplish this using affordable commercial GPS and orientation trackers, as well as computer vision, and simple and intuitive user input. We also present an1d evaluate different depth cue aids to help users in annotation, and different input controls for annotation.

Continuous Multimodal Authentication Using Dynamic Bayesian Networks

We present an approach to fuse contextual information with weak biometric measurements from multiple modalities in order to classify an individual's identity over time.

Interactive Stereoscopic FogScreen: Towards a 3D Walkthrough Display

Stephen DiVerdi, Cha Lee

We are working towards a 3D walk-through display based on the FogScreen, an "immaterial" indoor 2D projection screen. We extend the basic 2D FogScreen setup with head tracking to provide correct perspective rendering for a single user. Second, we add support for multiple types of stereoscopic imagery. Third, we present the front and back views of a graphics scene on the two sides of the FogScreen, so that one can cross the screen to see the content from the back.

Hand Gesture Recognition

Haiying Guan

Hand gestures provide abundant degrees of freedom (DOF) and are nature and intuitive for communication. Such advantages make hand gestures a potential input modality for 3D (or even higher) free-form input interfaces. We present an appearance-based 3D hand posture estimation method that determines a ranked set of possible hand posture candidates from a 2D unmarked hand image. We formulate the hand posture estimation problem as a many to many, nonlinear mapping problem in a high dimension space. A general algorithm called ISOSOM is proposed for nonlinear dimension reduction, and applied to 3D hand pose reconstruction to establish the mapping relationships between the hand poses and the image features.

Tangible Space Initiative

Sehyung Park, Chris Coffin

In collaboration with the Korea Institute for Science and Technology (KIST), we are working towards tangible interfaces for telecommunication. We present InViTe, the Interactive Video Teleconferencing system, which tackles integration of real and virtual imagery in a video-based augmented reality approach. Our goal is to provide a highly interactive tangible experience with augmented videos. We also present the first steps towards using haptic interfaces for easy graph manipulation.

Multi-Flash Stereopsis

Rogerio Feris, Longbin Chen

We describe a novel method for acquiring high quality depth maps from real-world scenes. Our approach is based on a compact stereo camera with small baseline multi-flash illumination, which allows discontinuity preserving correspondence maps from scenes that are extremely challenging for conventional stereo methods.

Facial Expression Analysis

Ya Chang

In order to automatically recognize and analyze human facial expression, we are working on probabilistic video-to-video facial expression recognition based on "manifold of facial expression." The manifold of facial expression is embedded in high dimensional image space. In the embedded space, images with different expressions can be clustered and classified by the probabilistic model learned on the manifold of expression.

Hand Tracking with Flocks of Features and Gesture based Game Interface

Jae-Sik Chang

Chris Coffin

We present a technique that allows a user to look beyond occluding objects in arbitrary 3D graphics scenes. In order to control this form of virtual x-ray vision, the user interactively cuts holes into the occluding geometry. The user can rapidly define a cutout shape or choose a standard shape and sweep it over the occluding wall segments to reveal what lies behind them.

Image-space correction of AR registration errors using graphics hardware

Steve DiVerdi

We present a hardware-accelerated image-based post-processing technique that adjusts rendering of virtual geometry to better match edges present in images of a physical scene, reducing the visual effect of registration errors from both inaccurate tracking and oversimplified modeling.

3D Models from a Single View Image

Taehee Lee, Brynjar Gretarsson

We implemented a method of reconstructing a 3D scene from a single perspective view image. The image may be an arbitrary scene of outdoor that contains the ground plane and some urban structures like buildings and walls. In order to get the 3D geometry, users interactively set a reference unit square on the ground plane, a vertical vanishing point, and a reference height. We are working towards acquiring rough 3D geometry from live video feeds in real time.