We investigated the computational properties of natural object hierarchy in the context of constellation object class models, and its utility for object class recognition. We first observed an interesting computational property of the object hierarchy: comparing the recognition rate when using models of objects at different levels, the higher more inclusive levels (e.g., Closed-Frame Vehicles or Vehicles) exhibit higher recall but lower precision when compared with the class specific level (e.g., bus). These inherent differences suggest that combining object classifiers from different hierarchical levels into a single classifier may improve classification, as it appears like these models capture different aspects of the object. We describe a method to combine these classifiers, and analyze the conditions under which improvement can be guaranteed. When given a small sample of a new object class, we describe a method to transfer knowledge across the tree hierarchy, between related objects. Finally, we describe extensive experiments using object hierarchies obtained from publicly available datasets, and show that the combined classifiers significantly improve recognition results.

Unexpected stimuli are a challenge to any machine learning algorithm. Here, we identify distinct types of unexpected events when general-level and specific-level classifiers give conflicting predictions. We define a formal framework for the representation and processing of incongruent events: Starting from the notion of label hierarchy, we show how partial order on labels can be deduced from such hierarchies. For each event, we compute its probability in different ways, based on adjacent levels in the label hierarchy. An incongruent event is an event where the probability computed based on some more specific level is much smaller than the probability computed based on some more general level, leading to conflicting predictions. Algorithms are derived to detect incongruent events from different types of hierarchies, different applications, and a variety of data types. We present promising results for the detection of novel visual and audio objects, and new patterns of motion in video. We also discuss the detection of Out-Of- Vocabulary words in speech recognition, and the detection of incongruent events in a multimodal audiovisual scenario.

Categorization is one of the fundamental building blocks of cognitive systems. Object categorization has traditionally been addressed in the vision domain, even though cognitive agents are intrinsically multimodal. Indeed, biological systems combine several modalities in order to achieve robust categorization. In this paper we propose a multimodal approach to object category detection, using audio and visual information. The auditory channel is modeled on biologically motivated spectral features via a discriminative classifier. The visual channel is modeled by a state of the art part based model. Multimodality is achieved using two fusion schemes, one high level and the other low level. Experiments on six different object categories, under increasingly difficult conditions, show strengths and weaknesses of the two approaches, and clearly underline the open challenges for multimodal category detection.IDIAP-RR 07-58 3

It is of prime importance in everyday human life to cope with and respond appropriately to events that are not foreseen by prior experience. Machines to a large extent lack the ability to respond appropriately to such inputs. An important class of unexpected events is defined by incongruent combinations of inputs from different modalities and therefore multimodal information provides a crucial cue for the identification of such events, e.g., the sound of a voice is being heard while the person in the field-of-view does not move her lips. In the project DIRAC ("Detection and Identification of Rare Audio-visual Cues") we have been developing algorithmic approaches to the detection of such events, as well as an experimental hardware platform to test it. An audio-visual platform ("AWEAR" - audio-visual wearable device) has been constructed with the goal to help users with disabilities or a high cognitive load to deal with unexpected events. Key hardware components include stereo panoramic vision sensors and 6-channel worn-behind-the-ear (hearing aid) microphone arrays. Data have been recorded to study audio-visual tracking, a/v scene/object classification and a/v detection of incongruencies. © 2008 ACM

Sharing information among multiple learning agents can accelerate learning. It could be particularly useful if learners operate in continuously changing environments, because a learner could benefit from previous experience of another learner to adapt to their new environment. Such group-adaptive learning has numerous applications, from predicting financial time-series, through content recommendation systems, to visual understanding for adaptive autonomous agents. Here we address the problem in the context of online adaptive learning. We formally define the learning settings of Group Online Adaptive Learning and derive an algorithm named Shared Online Adaptive Learning (SOAL) to address it. SOAL avoids explicitly modeling changes or their dynamics, and instead shares information continuously. The key idea is that learners share a common small pool of experts, which they can use in a weighted adaptive way. We define group adaptive regret and prove that SOAL maintains known bounds on the adaptive regret obtained for single adaptive learners. Furthermore, it quickly adapts when learning tasks are related to each other. We demonstrate the benefits of the approach for two domains: vision and text. First, in the visual domain, we study a visual navigation task where a robot learns to navigate based on outdoor video scenes. We show how navigation can improve when knowledge from other robots in related scenes is available. Second, in the text domain, we create a new dataset for the task of assigning submitted papers to relevant editors. This is, inherently, an adaptive learning task due to the dynamic nature of research fields evolving in time. We show how learning to assign editors improves when knowledge from other editors is available. Together, these results demonstrate the benefits for sharing information across learners in concurrently changing environments.