Compact Trilinear Interaction for Visual Question Answering
This addresses the computational complexity and memory requirements in VQA for researchers and practitioners, though it is incremental as it builds on existing trilinear and bilinear interaction methods.
The paper tackles the problem of efficiently modeling interactions between image, question, and answer inputs in Visual Question Answering by proposing a compact trilinear interaction model with tensor decomposition and knowledge distillation, achieving state-of-the-art results on TDIUC, VQA-2.0, and Visual7W datasets.
In Visual Question Answering (VQA), answers have a great correlation with question meaning and visual contents. Thus, to selectively utilize image, question and answer information, we propose a novel trilinear interaction model which simultaneously learns high level associations between these three inputs. In addition, to overcome the interaction complexity, we introduce a multimodal tensor-based PARALIND decomposition which efficiently parameterizes trilinear interaction between the three inputs. Moreover, knowledge distillation is first time applied in Free-form Opened-ended VQA. It is not only for reducing the computational cost and required memory but also for transferring knowledge from trilinear interaction model to bilinear interaction model. The extensive experiments on benchmarking datasets TDIUC, VQA-2.0, and Visual7W show that the proposed compact trilinear interaction model achieves state-of-the-art results when using a single model on all three datasets.