Augmented Attribute Representations

By Viktoriia Sharmanska, Novi Quadrianto, Christoph H. Lampert. 2012

90 citations recorded by the source corpus.

Subfields: Artificial Intelligence, 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 Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Transductive Multi-View Zero-Shot Learning (2015, methodology): “Zero-shot learning Multi-class single label zero-shot learning has now been widely studied using attribute-based intermediate semantic layers [3, 6, 10, 11, 19, 25] or data-driven [8, 9, 21, 30] representations.”
Learning Like a Child: Fast Novel Visual Concept Learning from Sentence Descriptions of Images (2015, methodology): “In addition, [47] adopted auto-encoders with attribute representations to learn new class labels and [56] proposed a method that scales to large datasets using label embeddings.”
Domain Adaptation for Visual Applications: A Comprehensive Survey (2017, background): “These attributes can be a set of nameable image characteristics and semantic concepts [84, 130, 158, 88], or latent topics discovered by the system from the data [199, 131, 89].”
Label-Embedding for Image Classification (2015, neutral): “While this extends DAP to learning with labeled data, this approach does not improve DAP for zero-shot recognition.”
Discovering Attribute Shades of Meaning with the Crowd (2015, background): “…shades based on image clustering—or non-semantic attribute discovery such as (Parikh and Grauman, 2011a; Mahajan et al, 2011; Duan et al, 2012; Rastegari et al, 2012; Sharmanska et al, 2012; Yu et al, 2013)—will be hard pressed to group images according to these perceived, possibly subtle, cues.”
Zero-shot recognition with unreliable attributes (2014, methodology): “Acknowledging that attribute classiﬁers are often unreliable, recent work abandons purely semantic attributes in favor of discovering mid-level features that are both detectable and discriminative for a set of class labels [11, 22, 26, 15, 30, 27, 1].”
Designing Category-Level Attributes for Discriminative Visual Recognition (2013, background): “[24] extends a semantic attribute representation with extra non-interpretable dimensions for enhanced discrimination.”
COSTA: Co-Occurrence Statistics for Zero-Shot Classification (2014, background): “For example, by including learned non-semantic attributes for better discrimination [19, 25], or to learn an attribute embedding specific for zero-shot prediction [1].”
Label-Embedding for Attribute-Based Classification (2013, background): “We now review related work on attributes, zero-shot learning and label embedding (three research areas which strongly overlap) with an emphasis on the latter.”
Transfer Learning in a Transductive Setting (2013, background): “We thus improve the graph structure by replacing the noisy raw input space with the more compact semantic output space which has shown to improve recognition [26, 22].”
NEIL: Extracting Visual Knowledge from Web Data (2013, background): “Our work is also related to attribute discovery [33] [35] since these approaches jointly discover the attributes and relationships between objects and attributes simultaneously.”
Attribute-Based Classification for Zero-Shot Visual Object Categorization (methodology): “Sharmanska et al. [37] propose an unsupervised technique for augmenting existing attribute representations with additional non-semantic binary features in order to make them more discriminative .”
Large-Scale Object Classification Using Label Relation Graphs (2014, methodology): “Finally, by modeling the label relations and ensuring consistency between visual predictions and semantic relations, our approach relates to work in transfer learning [31,30], zero-shot learning [28,12,25], and attribute-based recognition [1,36,33,14], especially those that use semantic knowledge to”
Learning Decision Trees Recurrently Through Communication (2021, methodology): “Attributes have been used for image description [14, 6], caption generation [47], face recognition [7], image retrieval [33, 54], action recognition [67, 63], novelty detection [57] and object classification [35, 53, 42, 52, 8].”