Technical Report for Ego4D Long-Term Action Anticipation Challenge 2025

TL;DR

This paper introduces a three-stage framework leveraging foundation models for long-term action anticipation in egocentric videos, achieving state-of-the-art results at CVPR 2025.

Contribution

The novel framework combines visual encoding, Transformer-based recognition, and LLM-driven anticipation, setting new benchmarks in long-term egocentric action prediction.

Findings

Achieved first place in the Ego4D LTA Challenge 2025

Established a new state-of-the-art in long-term action anticipation

Demonstrated effectiveness of combining visual features with large language models

Abstract

In this report, we present a novel three-stage framework developed for the Ego4D Long-Term Action Anticipation (LTA) task. Inspired by recent advances in foundation models, our method consists of three stages: feature extraction, action recognition, and long-term action anticipation. First, visual features are extracted using a high-performance visual encoder. The features are then fed into a Transformer to predict verbs and nouns, with a verb-noun co-occurrence matrix incorporated to enhance recognition accuracy. Finally, the predicted verb-noun pairs are formatted as textual prompts and input into a fine-tuned large language model (LLM) to anticipate future action sequences. Our framework achieves first place in this challenge at CVPR 2025, establishing a new state-of-the-art in long-term action prediction. Our code will be released at https://github.com/CorrineQiu/Ego4D-LTA-Challenge-2025.