Han-Pang Chiu

Virtual Training for Multi-View Object Class Recognition

Our goal is to circumvent one of the roadblocks to using existing approaches for single-view recognition for achieving multi-view recognition, namely, the need for sufficient training data for many viewpoints. We show how to construct virtual training examples for multi-view recognition using a simple model of objects (nearly planar facades centered at fixed 3D positions). We also show how the models can be learned from a few labeled images for each class.

Robust vision-aided navigation using Sliding-Window Factor graphs

This paper proposes a navigation algorithm that provides a low-latency solution while estimating the full nonlinear navigation state. Our approach uses Sliding-Window Factor Graphs, which extend existing incremental smoothing methods to operate on the subset of measurements and states that exist inside a sliding time window. We split the estimation into a fast short-term smoother, a slower but fully global smoother, and a shared map of 3D landmarks. A novel three-stage visual feature model is presented that takes advantage of both smoothers to optimize the 3D landmark map, while minimizing the computation required for processing tracked features in the short-term smoother. This three-stage model is formulated based on the maturity of the estimation of the 3D location of the underlying landmark in the map. Long-range associations are used as global measurements from matured landmarks in the short-term smoother and loop closure constraints in the long-term smoother. Experimental results demonstrate our approach provides highly-accurate solutions on large-scale real data sets using multiple sensors in GPS-denied settings.

2D Z-string: a new spatial knowledge representation for image databases

The knowledge structure called the 2D C+-string, proposed by Huang et al., to represent symbolic pictures allows a natural way to construct iconic indexes for images. According to the cutting mechanism of the 2D C+-string, an object may be partitioned into several subparts The number of partitioned subparts is bounded to O(n2), where n is the number of objects in the image. Hence, the string length is also bounded to O(n2). In this paper, we propose a new spatial knowledge representation called the 2D Z-string. Since there are no cuttings between objects in the 2D Z-string, the integrity of objects is preserved and the string length is bounded to O(n). Finally, some experiments are conducted to compare the performance of both approaches.

Stable vision-aided navigation for large-area augmented reality

In this paper, we present a unified approach for a drift-free and jitter-reduced vision-aided navigation system. This approach is based on an error-state Kalman filter algorithm using both relative (local) measurements obtained from image based motion estimation through visual odometry, and global measurements as a result of landmark matching through a pre-built visual landmark database. To improve the accuracy in pose estimation for augmented reality applications, we capture the 3D local reconstruction uncertainty of each landmark point as a covariance matrix and implicity rely more on closer points in the filter. We conduct a number of experiments aimed at evaluating different aspects of our Kalman filter framework, and show our approach can provide highly-accurate and stable pose both indoors and outdoors over large areas. The results demonstrate both the long term stability and the overall accuracy of our algorithm as intended to provide a solution to the camera tracking problem in augmented reality applications.

Constrained Optimal Selection for Multi-Sensor Robot Navigation Using Plug-and-Play Factor Graphs

This paper proposes a real-time navigation approach that is able to integrate many sensor types while fulfilling performance needs and system constraints. Our approach uses a plug-and-play factor graph framework, which extends factor graph formulation to encode sensor measurements with different frequencies, latencies, and noise distributions. It provides a flexible foundation for plug-and-play sensing, and can incorporate new evolving sensors. A novel constrained optimal selection mechanism is presented to identify the optimal subset of active sensors to use, during initialization and when any sensor condition changes. This mechanism constructs candidate subsets of sensors based on heuristic rules and a ternary tree expansion algorithm. It quickly decides the optimal subset among candidates by maximizing observability coverage on state variables, while satisfying resource constraints and accuracy demands. Experimental results demonstrate that our approach selects subsets of sensors to provide satisfactory navigation solutions under various conditions, on large-scale real data sets using many sensors.

Mining Structure Fragments for Smart Bundle Adjustment

© 2014. The copyright of this document resides with its authors. It may be distributed unchanged freely in print or electronic forms.

3D C-string: a new spatio-temporal knowledge representation for video database systems

Abstract In video database systems, one of the most important methods for discriminating the videos is by using the objects and the perception of spatial and temporal relations that exist between objects in the desired videos. In this paper, we propose a new spatio-temporal knowledge representation called 3D C-string. The knowledge structure of 3D C-string, extended from the 2D C+-string, uses the projections of objects to represent spatial and temporal relations between the objects in a video. Moreover, it can keep track of the motions and size changes of the objects in a video. The string generation and video reconstruction algorithms for the 3D C-string representation of video objects are also developed. By introducing the concept of the template objects and nearest former objects, the string generated by the string generation algorithm is unique for a given video and the video reconstructed from a given 3D C-string is unique too. This approach can provide us an easy and efficient way to retrieve, visualize and manipulate video objects in video database systems. Finally, some experiments are performed to show the performance of the proposed algorithms.

High-precision localization using visual landmarks fused with range data

Visual landmark matching with a pre-built landmark database is a popular technique for localization. Traditionally, landmark database was built with visual odometry system, and the 3D information of each visual landmark is reconstructed from video. Due to the drift of the visual odometry system, a global consistent landmark database is difficult to build, and the inaccuracy of each 3D landmark limits the performance of landmark matching. In this paper, we demonstrated that with the use of precise 3D Li-dar range data, we are able to build a global consistent database of high precision 3D visual landmarks, which improves the landmark matching accuracy dramatically. In order to further improve the accuracy and robustness, landmark matching is fused with a multi-stereo based visual odometry system to estimate the camera pose in two aspects. First, a local visual odometry trajectory based consistency check is performed to reject some bad landmark matchings or those with large errors, and then a kalman filtering is used to further smooth out some landmark matching errors. Finally, a disk-cache-mechanism is proposed to obtain the real-time performance when the size of the landmark grows for a large-scale area. A week-long real time live marine training experiments have demonstrated the high-precision and robustness of our proposed system.

Learning to generate novel views of objects for class recognition

Multi-view object class recognition can be achieved using existing approaches for single-view object class recognition, by treating different views as entirely independent classes. This strategy requires a large amount of training data for many viewpoints, which can be costly to obtain. We describe a method for constructing a weak three-dimensional model from as few as two views of an object of the target class, and using that model to transform images of objects from one view to several other views, effectively multiplying their value for class recognition. Our approach can be coupled with any 2D image-based recognition system. We show that automatically transformed images dramatically decrease the data requirements for multi-view object class recognition.

3D Z-string: A new knowledge structure to represent spatio-temporal relations between objects in a video

In this paper, we propose a new knowledge structure called 3D Z-string, extended from the 2D Z-string, to represent the spatial and temporal relations between objects in a video and to keep track of the motions and size changes of the objects. Since there are no cuttings between objects in the 3D Z-string, the integrity of objects is preserved. The string generation and video reconstruction algorithms for the 3D Z-string representation of video objects are also developed. The string generated by the string generation algorithm is unique for a given video and the video reconstructed from a given 3D Z-string is unique too. The experimental results show that the 3D Z-string is more compact and efficient than the 3D C-string in terms of storage space and execution time.