MemeTector: Enforcing deep focus for meme detection
Christos Koutlis, Manos Schinas, Symeon Papadopoulos

TL;DR
MemeTector introduces a novel deep learning approach that emphasizes critical image parts and uses attention mechanisms to improve meme detection accuracy and robustness, outperforming existing methods.
Contribution
The paper proposes Visual Part Utilization and an attention-enhanced ViT model to focus on key meme features, advancing meme detection techniques.
Findings
Light visual part utilization with sufficient text improves robustness.
The model surpasses state-of-the-art performance.
Training with controlled text presence enhances accuracy.
Abstract
Image memes and specifically their widely-known variation image macros, is a special new media type that combines text with images and is used in social media to playfully or subtly express humour, irony, sarcasm and even hate. It is important to accurately retrieve image memes from social media to better capture the cultural and social aspects of online phenomena and detect potential issues (hate-speech, disinformation). Essentially, the background image of an image macro is a regular image easily recognized as such by humans but cumbersome for the machine to do so due to feature map similarity with the complete image macro. Hence, accumulating suitable feature maps in such cases can lead to deep understanding of the notion of image memes. To this end, we propose a methodology, called Visual Part Utilization, that utilizes the visual part of image memes as instances of the regular…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsMisinformation and Its Impacts · Sentiment Analysis and Opinion Mining · Humor Studies and Applications
