ClipSitu: Effectively Leveraging CLIP for Conditional Predictions in Situation Recognition

TL;DR

This paper introduces ClipSitu, a model leveraging CLIP's visual-linguistic knowledge with enhanced MLP and cross-attention Transformer architectures to significantly improve situation recognition accuracy in images.

Contribution

It proposes a novel approach combining CLIP features with advanced MLP and cross-attention Transformer models for better situation recognition.

Findings

Outperforms state-of-the-art models by 14.1% in semantic role labeling accuracy.

Achieves superior situation localization performance.

Demonstrates the effectiveness of CLIP-based features in structured image understanding.

Abstract

Situation Recognition is the task of generating a structured summary of what is happening in an image using an activity verb and the semantic roles played by actors and objects. In this task, the same activity verb can describe a diverse set of situations as well as the same actor or object category can play a diverse set of semantic roles depending on the situation depicted in the image. Hence a situation recognition model needs to understand the context of the image and the visual-linguistic meaning of semantic roles. Therefore, we leverage the CLIP foundational model that has learned the context of images via language descriptions. We show that deeper-and-wider multi-layer perceptron (MLP) blocks obtain noteworthy results for the situation recognition task by using CLIP image and text embedding features and it even outperforms the state-of-the-art CoFormer, a Transformer-based model, thanks to the external implicit visual-linguistic knowledge encapsulated by CLIP and the expressive power of modern MLP block designs. Motivated by this, we design a cross-attention-based Transformer using CLIP visual tokens that model the relation between textual roles and visual entities. Our cross-attention-based Transformer known as ClipSitu XTF outperforms existing state-of-the-art by a large margin of 14.1\% on semantic role labelling (value) for top-1 accuracy using imSitu dataset. {Similarly, our ClipSitu XTF obtains state-of-the-art situation localization performance.} We will make the code publicly available.