Bottom-Up and Top-Down Attention for Image Captioning and Visual Question Answering

Top-down visual attention mechanisms have been used extensively in image captioning and visual question answering (VQA) to enable deeper image understanding through fine-grained analysis and even multiple steps of reasoning. In this work, we propose a combined bottom-up and top-down attention mechanism that enables attention to be calculated at the level of objects and other salient image regions. This is the natural basis for attention to be considered. Within our approach, the bottom-up mechanism (based on Faster R-CNN) proposes image regions, each with an associated feature vector, while the top-down mechanism determines feature weightings. Applying this approach to image captioning, our results on the MSCOCO test server establish a new state-of-the-art for the task, achieving CIDEr / SPICE / BLEU-4 scores of 117.9, 21.5 and 36.9, respectively. Demonstrating the broad applicability of the method, applying the same approach to VQA we obtain first place in the 2017 VQA Challenge.

PDF Abstract CVPR 2018 PDF CVPR 2018 Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Visual Question Answering GQA Test2019 BottomUp Accuracy 49.74 # 100
Binary 66.64 # 101
Open 34.83 # 102
Consistency 78.71 # 103
Plausibility 84.57 # 60
Validity 96.18 # 74
Distribution 5.98 # 57
Visual Question Answering VQA v2 test-std Up-Down overall 70.34 # 54

Methods


No methods listed for this paper. Add relevant methods here