Conservative Visual Learning for Object Detection with Minimal Hand Labeling Effort

Roth, Peter M.; Gräbner, Helmut; Skočaj, Danijel; Bischof, Horst; Leonardis, Aleš

doi:10.1007/11550518_37

Cited by 10 publications

(11 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The project comprises of work on architectures, environment representation, object recognition, human-machine interaction and natural language understanding, planning, or action representations (see e.g. Roth et al 2005;Meier et al 2006;Kruijff et al 2006)). One of the key ideas of this project is to integrate work on the different fields of AI and robotics research as well as the cognitive sciences.…”

Section: Recent (Cognitive) Robotics and Agent Applicationsmentioning

confidence: 99%

Robot Controllers for Highly Dynamic Environments with Real-time Constraints

Ferrein

2010

Künstl Intell

View full text Add to dashboard Cite

Section: Recent (Cognitive) Robotics and Agent Applicationsmentioning

confidence: 99%

Robot Controllers for Highly Dynamic Environments with Real-time Constraints

Ferrein

2010

Künstl Intell

View full text Add to dashboard Cite

“…Roth et al developed an online learning system for the task of person detection on surveillance camera images [14]. The system employs a reconstructive model using incremental principal component analysis for autonomously selecting positive examples for an online AdaBoost classifier.…”

Section: Related Workmentioning

confidence: 99%

“…Although the complete dimensionality of a single view representation x i is thus (50+3)x18x18=17172, the effective dimensionality is much smaller, due to the sparsity of the representation vector and the confinement of activation to the figure-ground mask. Nevertheless it is a key feature of our biologically motivated visual processing model that robustness, generalization and speed of learning is not achieved by a dimension reduction as in most other current online learning models [8,9,3,10,11,12,14]. The key element is a transformation of the input into a sparse robust feature map representation that captures relevant locally invariant structures of the objects.…”

Section: Hierarchical Feature Processingmentioning

confidence: 99%

Online Learning of Objects in a Biologically Motivated Visual Architecture

Wersing

Kirstein

Götting

et al. 2007

Int. J. Neur. Syst.

View full text Add to dashboard Cite

We present a biologically motivated architecture for object recognition that is capable of online learning of several objects based on interaction with a human teacher. The system combines biological principles such as appearance-based representation in topographical feature detection hierarchies and context-driven transfer between different levels of object memory. Training can be performed in an unconstrained environment by presenting objects in front of a stereo camera system and labeling them by speech input. The learning is fully online and thus avoids an artificial separation of the interaction into training and test phases. We demonstrate the performance on a challenging ensemble of 50 objects.

show abstract

“…Very recently we presented a preliminary version of the proposed approach [17]. It was based on batch methods, so it was not suitable for on-line learning.…”

Section: Introductionmentioning

confidence: 99%

On-line Conservative Learning for Person Detection

Roth

Gräbner

Skočaj

et al. 2005

2005 IEEE International Workshop on Visual Surveillance and Performance Evaluation of Tracking and Surveillance

View full text Add to dashboard Cite

We present a novel on-line conservative learningframework for an object detection system. All algorithms operate in an on-line mode, in particular we also present a novel on-line AdaBoost method. The basic idea is to start with a very simple object detection system and to exploit a huge amount of unlabeled video data by being very conservative in selecting training examples. The key idea is to use reconstructive and discriminative classifiers in an iterative co-training fashion to arrive at increasingly better object detectors. We demonstrate the framework on a surveillance task where we learn person detectors that are tested on two surveillance video sequences. We start with a simple moving object classifier and proceed with incremental PCA (on shape and appearance) as a reconstructive classifier which in turn generates a training setfor a discriminative on-line AdaBoost classifier

show abstract

Conservative Visual Learning for Object Detection with Minimal Hand Labeling Effort

Cited by 10 publications

References 19 publications

Robot Controllers for Highly Dynamic Environments with Real-time Constraints

Robot Controllers for Highly Dynamic Environments with Real-time Constraints

Online Learning of Objects in a Biologically Motivated Visual Architecture

On-line Conservative Learning for Person Detection

Contact Info

Product

Resources

About