On the relationship between visual attributes and convolutional networks

93 citations recorded by the source corpus.

How researchers see this paper

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

Multimodal Explanations: Justifying Decisions and Pointing to the Evidence (2018, methodology): “In this paradigm, models like [17] which highlight discriminative image attributes without attempting to model the classifiers reasoning process are considered justification explanations, whereas models like [37, 12, 39] which aim to illuminate the inner reasoning process of deep networks are consid”
Attentive Explanations: Justifying Decisions and Pointing to the Evidence (2016, methodology): “Some methods ﬁnd discriminative visual patches [12, 7] whereas others aim to understand intermediate features which are important for end decisions [40, 14, 41] e.g. what does a certain neuron represent.”
Improving Semantic Embedding Consistency by Metric Learning for Zero-Shot Classiffication (2016, background): “More recently, [10] showed the existence of Attribute Centric Nodes (ACNs) within CNNs trained to recognize objects, collectively encoding information pertinent to visual attributes, unevenly and sparsely distributed across all the layers of the network.”
Learning Attributes Equals Multi-Source Domain Generalization (2016, methodology): “recent work has applied deep models to attribute detection [11, 83, 47, 16].”
Image Captioning with Semantic Attention (2016, background): “This is motivated by the recent success of deep learning models on visual recognition tasks [10, 18].”