Q-learning of sequential attention for visual object recognition from informative local descriptors

By Lucas Paletta, Gerald Fritz, Christin Seifert. 2005

67 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

On the Binding Problem in Artificial Neural Networks (2020, methodology): “A policy for the agent (frequently implemented by a neural network) can then be evolved (Stanley and Miikkulainen, 2004), trained with Q-learning (Paletta et al., 2005), or via Policy Gradients (Butko and Movellan, 2009).”
Recurrent Models of Visual Attention (2014, background): “[2, 4, 6, 14, 16, 17, 20] have embraced vision as a sequential decision task as we do here.”
Attention mechanism in neural networks: where it comes and where it goes (2022, background): “An interesting study proposes to learn sequential attention in real-world visual object recognition using a Q-learner [36].”
Learning to Track: Online Multi-object Tracking by Decision Making (2015, background): “Markov decision processes [6] have been applied to different computer vision tasks, such as feature selection for recognition [35, 17], human activity forecasting [21], video game playing [28] and human-machine collaboration [37].”
Learning to combine foveal glimpses with a third-order Boltzmann machine (2010, methodology): “SIFT features have been proposed either as a prefilter for reducing the number of possible fixation positions [17] or as a way of preprocessing the raw glimpses [13].”
50 Years of object recognition: Directions forward (2013, background): “have published a number of papers proposing the use of reinforcement learning techniques — based on Q-learning — for deciding the image sub-regions where the recognition module should attend to and extract features from and achieve recognition [229], [230].”
Closed-Loop Learning of Visual Control Policies (2011, unknown): “Reinforcement learning can also be used to learn strategies for view selection (Paletta & Pinz, 2000) and sequential attention models ( Paletta, Fritz, & Seifert, 2005 ).”
Reinforcement Learning for Visual Object Detection (2016, background): “[25] and Butko and Movellan[3] developed remarkable early sequential models based on POMDPs for recognition and face detection.”
Perception and Developmental Learning of Affordances in Autonomous Robots (2007, background): “, in [19], selecting foci of attention for optimal integration of visual information in sequential object recognition, or in sequential object recognition [20].”
State-of-the-Art in Visual Attention Modeling (2013, unknown): “[164] 2005 - + - + - - + - - SIFT R DR COIL-20, TSG-20”