AIDIA - Adaptive Interface for Display InterAction

Stenger, Björn; Woodley, Thomas; Kim, Taekyun; Hernández, C.; Cipolla, Roberto

doi:10.5244/c.22.78

Cited by 7 publications

(7 citation statements)

References 23 publications

(34 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The proposed method can also be applied to an LDA mixture model [30] as in [7], or other LDA variants including direct LDAs [29] if they are piecewise linear models and are based on the Rayleigh quotient. See [22] for the application of the three-step update algorithm to the OSM for set-based object recognition.…”

Section: Object Cateogorisation By Caltech101 Datasetmentioning

confidence: 99%

“…The method of using the sufficient spanning set for the three steps, the component analysis of the two matrices in the numerator and the denominator, respectively, and for the discriminant component computations, allows for efficient incremental learning. Note that the number of input vectors for the numerator matrix in many methods such as the Oriented Component Analysis (OCA) [5] and Orthogonal Subspace Method (OSM) [13,22] criteria, is often large in practice. In these cases the previous incremental LDA algorithms suffer due to the assumption of a small number of input vectors for the scatter matrix in the numerator (e.g.…”

Section: Object Cateogorisation By Caltech101 Datasetmentioning

confidence: 99%

See 1 more Smart Citation

Incremental Linear Discriminant Analysis Using Sufficient Spanning Sets and Its Applications

et al. 2010

View full text Add to dashboard Cite

This paper presents an incremental learning solution for Linear Discriminant Analysis (LDA) and its applications to object recognition problems. We apply the sufficient spanning set approximation in three steps i.e. update for the total scatter matrix, between-class scatter matrix and the projected data matrix, which leads an online solution which closely agrees with the batch solution in accuracy while significantly reducing the computational complexity. The algorithm yields an efficient solution to incremental LDA even when the number of classes as well as the set size is large. The incremental LDA method has been also shown useful for semi-supervised online learning. Label propagation is done by integrating the incremental LDA into an EM framework. The method has been demonstrated in the task of merging large datasets which were collected during MPEG standardization for face image retrieval, face authentication using the BANCA dataset, and object categorisation using the Caltech101 dataset.

show abstract

Section: Object Cateogorisation By Caltech101 Datasetmentioning

confidence: 99%

Section: Object Cateogorisation By Caltech101 Datasetmentioning

confidence: 99%

Incremental Linear Discriminant Analysis Using Sufficient Spanning Sets and Its Applications

et al. 2010

View full text Add to dashboard Cite

show abstract

“…The system includes an attention mechanism that allows one user at a time to be in control. Note that face recognition could be employed for customizing the interface, as done in a previous version of our system [74]. An active region is defined for the current user, within which the hand is tracked.…”

Section: Discussionmentioning

confidence: 99%

“…The combination of NCC and CM (NCC-CM-3, i.e. a threshold value of 0.3) that was proposed in [74] performs well in terms of precision, only slightly slightly worse compared to evaluating the same observers in parallel. Some combinations show higher precision, e.g.…”

Section: Observer Combinationsmentioning

confidence: 99%

A Vision-Based Remote Control

Stenger

Woodley

Cipolla

2010

Studies in Computational Intelligence

View full text Add to dashboard Cite

This chapter presents a vision-based system for touch-free interaction with a display at a distance. A single camera is fixed on top of the screen and is pointing towards the user. An attention mechanism allows the user to start the interaction and control a screen pointer by moving their hand in a fist pose directed at the camera. On-screen items can be chosen by a selection mechanism. Current sample applications include browsing video collections as well as viewing a gallery of 3D objects, which the user can rotate with their hand motion. We have included an up-to-date review of hand tracking methods, and comment on the merits and shortcomings of previous approaches. The proposed tracker uses multiple cues, appearance, color, and motion, for robustness. As the space of possible observation models is generally too large for exhaustive online search, we select models that are suitable for the particular tracking task at hand. During a training stage, various off-the-shelf trackers are evaluated. From this data different methods of fusing them online are investigated, including parallel and cascaded tracker evaluation. For the case of fist tracking, combining a small number of observers in a cascade results in an efficient algorithm that is used in our gesture interface. The system has been on public display at conferences where over a hundred users have engaged with it.

show abstract

“…As an example, if both observers returned a confidence value of 0.9, the expected error of NCC is lower than that of RT and NCC should be chosen. Color probability map, blob detection scale space maximum probability score M [22] Motion probability map, blob detection scale space maximum probability score CM [14] Color and motion probability map scale space maximum probability score …”

Section: Evaluating Multiple Observersmentioning

confidence: 99%

Learning to track with multiple observers

Stenger

Woodley

Cipolla

2009

2009 IEEE Conference on Computer Vision and Pattern Recognition

View full text Add to dashboard Cite

We propose a novel approach to designing algorithms for object tracking based on fusing multiple observation models. As the space of possible observation models is too large for exhaustive on-line search, this work aims to select models that are suitable for a particular tracking task at hand. During an off-line training stage observation models from various off-the-shelf trackers are evaluated. From this data different methods of fusing the observers on-line are investigated, including parallel and cascaded evaluation. Experiments on test sequences show that this evaluation is usefulfor automatically designing and assessing algorithms for a particular tracking task. Results are shown for face tracking with a handheld camera and hand tracking for gesture interaction. We show that for these cases combining a small number of observers in a sequential cascade results in efficient algorithms that are both robust and precise.

show abstract

AIDIA - Adaptive Interface for Display InterAction

Cited by 7 publications

References 23 publications

Incremental Linear Discriminant Analysis Using Sufficient Spanning Sets and Its Applications

Incremental Linear Discriminant Analysis Using Sufficient Spanning Sets and Its Applications

A Vision-Based Remote Control

Learning to track with multiple observers

Contact Info

Product

Resources

About