An Empirical Study on Leveraging Images in Automated Bug Report Reproduction

2025-02-20

AI Summary

Existing automated bug reproduction tools largely ignore image data—such as UI screenshots—in bug reports, relying solely on textual descriptions, videos, or logs. Method: This paper systematically identifies six functional roles of images in defect reports (e.g., state indication, interaction path illustration, anomaly visualization), empirically analyzes their distribution and multimodal (text–image) coordination patterns, and proposes techniques for image semantic–contextual association modeling, joint text–image pattern mining, and reproduction performance attribution analysis. Contribution/Results: We demonstrate that incorporating image information significantly improves reproduction accuracy (average +18.7%) and is irreplaceable by textual or log-based approaches alone. This work bridges a long-standing gap in bug reproduction research by establishing the critical, underexplored role of images, providing both theoretical foundations and actionable technical pathways for designing image-aware reproduction tools.

Technology Category

• Computer Vision: Image and Video Retrieval
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning
• Reasoning under Uncertainty: Uncertainty Representations

Application Category

• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

Automated bug reproduction is a challenging task, with existing tools typically relying on textual steps-to-reproduce, videos, or crash logs in bug reports as input. However, images provided in bug reports have been overlooked. To address this gap, this paper presents an empirical study investigating the necessity of including images as part of the input in automated bug reproduction. We examined the characteristics and patterns of images in bug reports, focusing on (1) the distribution and types of images (e.g., UI screenshots), (2) documentation patterns associated with images (e.g., accompanying text, annotations), and (3) the functional roles they served, particularly their contribution to reproducing bugs. Furthermore, we analyzed the impact of images on the performance of existing tools, identifying the reasons behind their influence and the ways in which they can be leveraged to improve bug reproduction. Our findings reveal several key insights that demonstrate the importance of images in supporting automated bug reproduction. Specifically, we identified six distinct functional roles that images serve in bug reports, each exhibiting unique patterns and specific contributions to the bug reproduction process. This study offers new insights into tool advancement and suggests promising directions for future research.

Problem

• Investigates the role of images in bug reports
• Analyzes image characteristics and their impact
• Demonstrates images' importance in bug reproduction

Innovation

• Leveraging images in bug reports
• Analyzing image functional roles
• Improving automated bug reproduction