Answer-Type Prediction for Visual Question Answering

By Kushal Kafle, Christopher Kanan. 2016

120 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

Stacking with Auxiliary Features for Visual Question Answering (2018, methodology): “The auxiliary features that we use are motivated by an analysis of the VQA dataset and also inspired by related work, such as using a Bayesian framework to predict the form of the answer from the question (Kaﬂe and Kanan, 2016).”
REMIND Your Neural Network to Prevent Catastrophic Forgetting (2019, methodology): “To adapt ExStream and SLDA for VQA, we use a variant of the linear VQA model in [60], which concatenates ResNet-101 image features to question features extracted from a universal sentence encoder [61] and then feeds them into a linear classiﬁer.”
Don't Just Assume; Look and Answer: Overcoming Priors for Visual Question Answering (2017, background): “has received a lot of attention – a number of VQA datasets have been curated [5, 21, 41, 12, 25, 11, 29, 13, 39] and a variety of deep-learning models have been developed [5, 9, 38, 37, 16, 3, 34, 19, 24, 4, 31, 20, 10, 26, 14, 35, 36, 40, 30].”
Visual question answering: Datasets, algorithms, and future challenges (2016, methodology): “ but many of them can be accurately answered without using the image due to language biases. Relatively simple image-blind algorithms have achieved 49.6% accuracy on COCO-VQA using the question alone [36]. The dataset also contains many subjective, opinion-seeking questions that do not have a single”
An Analysis of Visual Question Answering Algorithms (2017, methodology): “They were shown random images from COCO and Visual Genome and could also upload images.”
Structured Attentions for Visual Question Answering (2017, methodology): “There have been many different directions for improving VQA performance, including predicting answer types [14], utilizing task-speciﬁc submodules [1, 2], and better multimodal feature pooling methods [8, 17].”
Visual question answering: A survey of methods and datasets (2016, background): “[34] integrate an explicit prediction of the type of expected answer from the question and formulate the answering in a Bayesian framework.”
There is Strength in Numbers: Avoiding the Hypothesis-Only Bias in Natural Language Inference via Ensemble Adversarial Training (2020, background): “In visual question answering datasets, several studies found it is often possible to answer the question without looking at the corresponding image (Zhang et al., 2016; Kaﬂe and Kanan, 2016; Goyal et al., 2017; Agrawal et al., 2018).”
Answer Them All! Toward Universal Visual Question Answering Models (2019, methodology): “Many algorithms for natural image VQA have been proposed, including Bayesian approaches [23, 36], methods using spatial attention [52, 33, 40, 6], compositional approaches [7, 8, 18], bilinear pooling schemes [29, 14], and others [50, 41, 26].”
Data Augmentation for Visual Question Answering (2017, methodology): “Some notable formulations include attention based methods (Yang et al., 2016; Xiong et al., 2016; Lu et al., 2016; Fukui et al., 2016), Bayesian frameworks (Kaﬂe and Kanan, 2016; Malinowski and Fritz, 2014), and compositional approaches (Andreas et al., 2016a,b).”
Challenges and Prospects in Vision and Language Research (2019, methodology): “For example, blind VQA models that “guess” the answers without looking at images show relatively high accuracy (Kafle and Kanan, 2016).”
Making the V in VQA Matter: Elevating the Role of Image Understanding in Visual Question Answering (2016, methodology): “uilding trust for machines among their users. 2. Related Work Visual Question Answering. A number of recent works have proposed visual question answering datasets [3,21,25, 30,10,45,37,35] and models [9,24,2,42,23,26,46, 44,43,40,34,19,28,14,41,32,16]. Our work builds on top of the VQA dataset from ”
Can you fool AI with adversarial examples on a visual Turing test? (2017, background): “A few recent works have proposed to extract more information from the question, such as the question type [22] or integrating additional attention mechanism over the question [31].”
Video Fill in the Blank with Merging LSTMs (2016, background): “It’s also highly related to the Visual-Question-and-Answer (VQaA) problem [1, 4, 11, 2, 9, 16].”
EarthGPT: A Universal Multimodal Large Language Model for Multisensor Image Comprehension in Remote Sensing Domain (2024, methodology): “The typical specialist models on VQA includes Qonly [44], RSVQA [45], SAN [46], and MQVQA [16].”