Less Is More: Linear Layers on CLIP Features as Powerful VizWiz Model
Abstract
Current architectures for multi-modality tasks such as visual question answering suffer from their high complexity. As a result, these architectures are difficult to train and require high computational resources. To address these problems we present a CLIP-based architecture that does not require any fine-tuning of the feature extractors. A simple linear classifier is used on the concatenated features of the image and text encoder. During training an auxiliary loss is added which operates on the answer types. The resulting classification is then used as an attention gate on the answer class selection. On the VizWiz 2022 Visual Question Answering Challenge we achieve 60.15 % accuracy on Task 1: Predict Answer to a Visual Question and AP score of 83.78 % on Task 2: Predict Answerability of a Visual Question.
- Publication:
-
arXiv e-prints
- Pub Date:
- June 2022
- DOI:
- 10.48550/arXiv.2206.05281
- arXiv:
- arXiv:2206.05281
- Bibcode:
- 2022arXiv220605281D
- Keywords:
-
- Computer Science - Computer Vision and Pattern Recognition;
- Computer Science - Computation and Language;
- Computer Science - Machine Learning
- E-Print:
- VizWiz Grand Challenge: Describing Images and Videos Taken by Blind People (CVPR Workshop 2022)