Show, Ask, Attend, and Answer: A Strong Baseline For Visual Question Answering

11 Apr 2017Vahid KazemiAli Elqursh

This paper presents a new baseline for visual question answering task. Given an image and a question in natural language, our model produces accurate answers according to the content of the image... (read more)

PDF Abstract
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
Visual Question Answering VQA v1 test-dev SAAA (ResNet) Accuracy 64.5 # 1
Visual Question Answering VQA v1 test-std SAAA (ResNet) Accuracy 64.6 # 1

Methods used in the Paper