Saved in:
Bibliographic Details
Main Authors: Iqbal, Salman, Rehman, Waheed
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.08289
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910046332387328
author Iqbal, Salman
Rehman, Waheed
author_facet Iqbal, Salman
Rehman, Waheed
contents Zero-shot action recognition relies on transferring knowledge from vision-language models to unseen actions using semantic descriptions. While recent methods focus on temporal modeling or architectural adaptations to handle video data, we argue that semantic prompting alone provides a strong and underexplored signal for zero-shot action understanding. We introduce SP-CLIP, a lightweight framework that augments frozen vision-language models with structured semantic prompts describing actions at multiple levels of abstraction, such as intent, motion, and object interaction. Without modifying the visual encoder or learning additional parameters, SP-CLIP aligns video representations with enriched textual semantics through prompt aggregation and consistency scoring. Experiments across standard benchmarks show that semantic prompting substantially improves zero-shot action recognition, particularly for fine-grained and compositional actions, while preserving the efficiency and generalization of pretrained models.
format Preprint
id arxiv_https___arxiv_org_abs_2603_08289
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Novel Semantic Prompting for Zero-Shot Action Recognition
Iqbal, Salman
Rehman, Waheed
Computer Vision and Pattern Recognition
Zero-shot action recognition relies on transferring knowledge from vision-language models to unseen actions using semantic descriptions. While recent methods focus on temporal modeling or architectural adaptations to handle video data, we argue that semantic prompting alone provides a strong and underexplored signal for zero-shot action understanding. We introduce SP-CLIP, a lightweight framework that augments frozen vision-language models with structured semantic prompts describing actions at multiple levels of abstraction, such as intent, motion, and object interaction. Without modifying the visual encoder or learning additional parameters, SP-CLIP aligns video representations with enriched textual semantics through prompt aggregation and consistency scoring. Experiments across standard benchmarks show that semantic prompting substantially improves zero-shot action recognition, particularly for fine-grained and compositional actions, while preserving the efficiency and generalization of pretrained models.
title Novel Semantic Prompting for Zero-Shot Action Recognition
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2603.08289