pith. sign in

arxiv: 2109.04422 · v1 · pith:YUMTDN4Ynew · submitted 2021-09-09 · 💻 cs.CV · cs.CL

TxT: Crossmodal End-to-End Learning with Transformers

classification 💻 cs.CV cs.CL
keywords multimodalend-to-endtransformer-basedvisualdetectorslearningansweringcrossmodal
0
0 comments X
read the original abstract

Reasoning over multiple modalities, e.g. in Visual Question Answering (VQA), requires an alignment of semantic concepts across domains. Despite the widespread success of end-to-end learning, today's multimodal pipelines by and large leverage pre-extracted, fixed features from object detectors, typically Faster R-CNN, as representations of the visual world. The obvious downside is that the visual representation is not specifically tuned to the multimodal task at hand. At the same time, while transformer-based object detectors have gained popularity, they have not been employed in today's multimodal pipelines. We address both shortcomings with TxT, a transformer-based crossmodal pipeline that enables fine-tuning both language and visual components on the downstream task in a fully end-to-end manner. We overcome existing limitations of transformer-based detectors for multimodal reasoning regarding the integration of global context and their scalability. Our transformer-based multimodal model achieves considerable gains from end-to-end learning for multimodal question answering.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.