Less is More: Toward Zero-Shot Local Scene Graph Generation via Foundation Models

TL;DR

This paper introduces ELEGANT, a zero-shot framework for local scene graph generation using foundation models, which abstracts structural information with partial objects to enhance reasoning in downstream tasks.

Contribution

It proposes a novel zero-shot local scene graph generation task and a framework leveraging foundation models for perception and reasoning without labeled supervision.

Findings

Outperforms baselines in open-ended evaluation with ECLIPSE metric.

Achieves up to 24.58% improvement over prior methods in close-set setting.

Demonstrates strong reasoning capabilities of foundation models in structural understanding.

Abstract

Humans inherently recognize objects via selective visual perception, transform specific regions from the visual field into structured symbolic knowledge, and reason their relationships among regions based on the allocation of limited attention resources in line with humans' goals. While it is intuitive for humans, contemporary perception systems falter in extracting structural information due to the intricate cognitive abilities and commonsense knowledge required. To fill this gap, we present a new task called Local Scene Graph Generation. Distinct from the conventional scene graph generation task, which encompasses generating all objects and relationships in an image, our proposed task aims to abstract pertinent structural information with partial objects and their relationships for boosting downstream tasks that demand advanced comprehension and reasoning capabilities. Correspondingly, we introduce zEro-shot Local scEne GrAph geNeraTion (ELEGANT), a framework harnessing foundation models renowned for their powerful perception and commonsense reasoning, where collaboration and information communication among foundation models yield superior outcomes and realize zero-shot local scene graph generation without requiring labeled supervision. Furthermore, we propose a novel open-ended evaluation metric, Entity-level CLIPScorE (ECLIPSE), surpassing previous closed-set evaluation metrics by transcending their limited label space, offering a broader assessment. Experiment results show that our approach markedly outperforms baselines in the open-ended evaluation setting, and it also achieves a significant performance boost of up to 24.58% over prior methods in the close-set setting, demonstrating the effectiveness and powerful reasoning ability of our proposed framework.