Leveraging Vision-Language Pre-training for Human Activity Recognition in Still Images

Cristina Mahanta, Gagan Bhatia
6/16/2025
cs.CVcs.CL

No confident code implementation found for this paper. 1 possible match can be shown below.

Abstract

Recognising human activity in a single photo enables indexing, safety and assistive applications, yet lacks motion cues. Using 285 MSCOCO images labelled as walking, running, sitting, and standing, scratch CNNs scored 41% accuracy. Fine-tuning multimodal CLIP raised this to 76%, demonstrating that contrastive vision-language pre-training decisively improves still-image action recognition in real-world deployments.

View on arXivView PDF

Code Implementations

No confident code match yet

We couldn't find an author-owned or strongly-evidenced community implementation for this paper. 1 weaker match is hidden by default — verify before relying on them.

No code implementations found yet.

Know of an implementation? Let us know in the comments below!

Similar Papers

Advancing Vision-based Human Action Recognition: Exploring Vision-Language CLIP Model for Generalisation in Domain-Independent Tasks

82% match2025

An Animation-based Augmentation Approach for Action Recognition from Discontinuous Video

82% match2024

Is end-to-end learning enough for fitness activity recognition?

82% match2023

RNNs, CNNs and Transformers in Human Action Recognition: A Survey and a Hybrid Model

80% match2024

Future Aspects in Human Action Recognition: Exploring Emerging Techniques and Ethical Influences

80% match2024

Pose Matters: Evaluating Vision Transformers and CNNs for Human Action Recognition on Small COCO Subsets

79% match2025

Cite this paper

@article{mahanta2025leveraging,
  title  = {Leveraging Vision-Language Pre-training for Human Activity Recognition in Still Images},
  author = {Cristina Mahanta and Gagan Bhatia},
  year   = {2025},
  eprint = {2506.13458},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2506.13458v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2506.13458.svg)](https://codeofpaper.com/paper/2506.13458)

Discussion