SUN attribute database: Discovering, annotating, and recognizing scene attributes

By Genevieve Patterson, James Hays. 2012

937 citations recorded by the source corpus.

Subfields: 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 Computer Vision and Pattern Recognition

Citations in the corpus

Microsoft COCO: Common Objects in Context (2014, methodology): “Finally, our dataset could provide a good benchmark for other types of labels, including scene types [3], attributes [9], [8] and full sentence written descriptions [51].”
Describing Textures in the Wild (2013, background): “Recently visual attributes have raised significant interest in the community [6, 11, 17, 25].”
ELEVATER: A Benchmark and Toolkit for Evaluating Language-Augmented Visual Models (2022, neutral): “They are usually in single domain, with a manual split of categories to product disjoint training and test categories, e.g., Animal with Attributes (AwA) [31], Caltech-UCSD Birds-200 (CUB) [64], SUN [47], aPY [14], and ZS-ImageNet [52, 17].”
Visual7W: Grounded Question Answering in Images (2015, background): “The 7W questions roughly correspond to an array of standard vision tasks: what [9, 15, 39], where [24, 50], when [30, 32], who [35, 42], why [33], how [23, 31] and which [16, 17].”
Learning Deep Features for Discriminative Localization (2015, background): “3.2 (i.e., thresholding) to first identify bird bounding boxes in both the train and test sets.”
Rectification-based Knowledge Retention for Continual Learning (2021, background): “We experiment with four datasets for the task incremental generalized zero-shot learning (GZSL) problem i.e. Attribute Pascal and Yahoo (aPY) [6], Animals with Attributes 1 (AWA1) [31], Caltech-UCSD-Birds 200-2011 (CUB) [29], and SUN Attribute dataset (SUN) [19].”
Zero-Shot Learning—A Comprehensive Evaluation of the Good, the Bad and the Ugly (2017, methodology): “ing, we select two coarse-grained, one small (aPY [18]) and one medium-scale (AWA1 [1]), and two fine-grained, both medium-scale, datasets (SUN [16], CUB [17]) with attributes and one large-scale dataset (ImageNet [19]) without.”
Zero-Shot Learning — The Good, the Bad and the Ugly (2017, background): “We thoroughly evaluate the second aspect of zero-shot learning, by using multiple splits of several small to largescale datasets [28, 38, 22, 10, 9].”
An embarrassingly simple approach to zero-shot learning (2015, methodology): “In order to make our approach easily comparable with the state of the art, we used the set of standard features provided by the authors of the data [16, 21, 30], including SIFT [25], and PHOG [6].”
From generic to specific deep representations for visual recognition (2014, background): “non-ConvNet [35] [22] [40] [26] [38] [45] [25] [12] [18] [24] [42] [30] [37] [46] [37] [37] [4] 71.”
Learning Temporal Transformations from Time-Lapse Videos (2016, background): “Object state recognition: Previous works [4,5,6] have looked at the problem of recognizing attributes, which has signiﬁcant conceptual overlap with the idea of object state recognition.”
Places: An Image Database for Deep Scene Understanding (2016, methodology): “In the SUN Attribute experiment [27], the training set size is 150 images per attribute.”
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders (2018, methodology): “We evaluate our framework on four widely used benchmark datasets CUB-200-2011 [33] (CUB), SUN attribute (SUN) [21], Animals with Attributes 1 and 2 (AWA1 [15], AWA2 [35]) for the GZSL and GFSL settings.”