Zero-Shot Learning via Visual Abstraction

By Stanislaw Antol, C. Lawrence Zitnick, Devi Parikh. 2014

88 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

Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotations (2016, background): “Relationships have already shown their utility in improving visual cognitive tasks (Antol et al. 2014; Yang et al. 2012).”
VQA: Visual Question Answering (2015, methodology): “To attract researchers interested in exploring the high-level reasoning required for VQA, but not the low-level vision tasks, we create a new abstract scenes dataset [1, 48, 49, 50] containing 50K scenes.”
Learning to Reconstruct Shapes from Unseen Classes (2018, background): “More recently, many researchers have begun to study zero- or few-shot recognition with deep networks [Akata et al., 2016, Antol et al., 2014, Hariharan and Girshick, 2017, Wang et al., 2017, Wang and Hebert, 2016].”
Zero-Shot Learning—A Comprehensive Evaluation of the Good, the Bad and the Ugly (2017, background): “[31], [34], [50], [51], [52], [53], [54] which exploit other auxiliary information that reduces this annotation effort.”
Learning Like a Child: Fast Novel Visual Concept Learning from Sentence Descriptions of Images (2015, background): “Zero-shot and one-shot learning For zero-shot learning, the task is to associate dense word vectors or attributes with image features [49, 17, 12, 2, 31].”
Survey of the State of the Art in Natural Language Generation: Core tasks, applications and evaluation (2017, background): “Progress in zero-shot learning, where the aim is to identify or categorise images for which little or no training data is available, is likely to contribute to the resolution of data sparseness problems (e.g. Antol et al., 2014; Elhoseiny et al., 2017).”
Visual question answering: Datasets, algorithms, and future challenges (2016, background): “The other major shortcoming is that it only has four kinds of questions, and these are limited to the kinds of things described in COCO’s captions.”
Don't just listen, use your imagination: Leveraging visual common sense for non-visual tasks (2015, methodology): “Despite being synthetic, it has been shown that semantic concepts learnt from abstract scenes can generalize to real images [1].”
Factorizable Net: An Efficient Subgraph-based Framework for Scene Graph Generation (2018, background): “In the early stage, most of the works targeted on using speciﬁc types of visual relations, such as spatial relations [5, 10, 14, 20, 26, 30] and actions ( i . e . interactions between objects) [1, 9, 11, 16, 19, 46, 47, 50, 56, 57, 60].”
Crowdsourcing in Computer Vision (2016, unknown): “Since some actions may be difficult to name but easy to visualize, Antol et al. [2014] design an interface to create cartoon depictions of actions.”
Adaptive SVM+: Learning with Privileged Information for Domain Adaptation (2017, methodology): “Discussion of Results : A summary of the classiﬁcations results on the INTERACT dataset is presented in Table 2.”
Detecting Visual Relationships with Deep Relational Networks (2017, methodology): “In most of these studies, relationships are usually extracted using simple heuristics or hand-crafted features, and used as an auxiliary components to facilitate other tasks, such as object recognition [25–33], image classiﬁcation and retrieval [34,35], scene understanding 1 code available at…”
Measuring Machine Intelligence Through Visual Question Answering (2016, methodology): “To enable multiple avenues for researching VQA, we introduce abstract scenes into the dataset [2, 34, 35, 36].”
On-line simultaneous learning and recognition of everyday activities from virtual reality performances (2017, unknown): “Recent work in this area has focused a great deal on one- and zero-shot learning techniques [8][9], which seek to minimize both the volume of and reliance on training datasets, with the extreme case being systems capable of classifying previously unseen actions without any prior training.”
A Semantical & Analytical Approach for Zero Shot Gesture Learning (2017, unknown): “Literature includes the use of semantic features [9], [14]– [16], textual description of categories to build visual models [17], and manipulating visual abstract visualizations [18].”
Sherlock: Scalable Fact Learning in Images (2015, methodology): “We began our data collection by augmenting existing datasets with fact language view labels fl: PPMI [41], Stanford40 [42], Pascal Actions [11], Sports [16], Visual Phrases [34], INTERACT [4] datasets.”
Zero-Shot Activity Recognition with Verb Attribute Induction (2017, background): “rb-level classes (and for some, sufﬁcient data points per class). The abstractness of broad-coverage activity labels makes the problem much more difﬁcult to study with attributes. To get around this, Antol et al. (2014) present a synthetic dataset of cartooncharacters performingdyadicactions,anduse ”
A Survey of Zero-Shot Learning (2019, background): “For example, in [7], the problem is activity recognition in images.”