Enhancing Online Shopping for VIPs: Insights from the Revamp System
- Nilotpal Biswas
- Mar 3
- 3 min read
In our previous discussions, we presented the summary of significant challenges visually impaired people (VIP) face while shopping in offline stores, online platforms, and virtual reality (VR) environments. In this blog, we delve into a research article titled "Revamp: Enhancing the Accessible Information-Seeking Experience of Online Shopping for Blind or Low Vision Users," which addresses issues faced by the VIPs in online shopping.
The Challenges of Online Shopping for VIPs
Online shopping has revolutionized accessibility for many, yet VIPs continue to face significant barriers. Two primary challenges hinder their experience:
Inadequate Image Descriptions: Product images are often accompanied by poor or missing alt-text, preventing VIPs from accessing crucial visual details like color, shape, and size.
Difficulty Navigating Information: E-commerce websites contain vast amounts of text and cluttered interfaces, making it difficult for screen readers to extract meaningful information efficiently.
Due to these challenges, many VIPs rely on sighted assistance—whether from family members, friends, or crowd-sourced helpers—to make informed purchase decisions. However, this dependency is not always convenient or desirable.
Revamp: A Bridge Between Reviews and Accessibility
The authors of the paper developed a system called Revamp, a browser extension that enhances the online shopping experience for VIPs. They implemented the extension on Amazon.com, which simplifies product pages and uses customer reviews to provide meaningful descriptions and answers to visual questions.
Key Features of Revamp:
Enhanced Image Descriptions: Revamp extracts descriptive snippets from product reviews to generate meaningful image descriptions, helping VIPs visualize products better.
Review-Based Q&A System: The system allows users to ask questions about visual attributes like color, shape, and logo. It then retrieves relevant answers from customer reviews.
Sentiment Summaries: Revamp provides sentiment-based insights by classifying reviews into positive and negative lists, enabling users to understand customer opinions at a glance.
Simplified Page Structure: To improve screen reader efficiency, Revamp restructures product pages by removing unnecessary elements like advertisements and non-essential navigation options.
Research Findings and User Feedback
The research team conducted extensive studies to evaluate the effectiveness of Revamp. Key findings include:
Understanding VIP Shopping Needs: Through interviews with VIPs, researchers identified four critical visual aspects necessary for product comprehension—color, logo, shape, and size.
Rule-Based Review Extraction: The authors developed syntactic rules to extract informative review snippets, ensuring that only relevant details contribute to image descriptions and Q&A responses.
User Evaluation: Eight VIP participants tested Revamp across different product categories (Home & Kitchen, Clothing & Jewelry, and Electronics). Results showed that:
85% of the extracted review snippets were found to be useful for understanding product appearance.
Participants reported greater independence in shopping decisions and preferred Revamp’s structured review summaries over raw Amazon reviews.
The system significantly reduced the effort required to locate key product information.
How revamp is unique
Unlike existing solutions such as Seeing AI, which offers generic image descriptions, Revamp provides context-specific insights directly from customer reviews. Instead of describing an item as simply "a red shirt," Revamp offers richer details like "a bright red, soft cotton shirt with a classic round-neck design."
By integrating Natural Language Processing (NLP) and Human-Computer Interaction (HCI) techniques, Revamp offers a structured, interactive, and user-friendly experience that makes online shopping more accessible.
Future Possibilities
The researchers envision several enhancements for Revamp, including:
Expanding to More Retail Platforms: While currently implemented for Amazon, Revamp could be adapted for other e-commerce sites like eBay, Walmart, or specialized fashion stores.
Improving High-Level Concept Descriptions: Future iterations may incorporate machine learning models to infer additional product characteristics like texture, quality, and usability.
Integrating Voice Assistants: Compatibility with AI assistants like Alexa or Google Assistant could make Revamp more accessible for mobile users.
Conclusion
The Revamp system marks a significant step forward in making online shopping truly inclusive for VIPs. By addressing key accessibility challenges through an innovative, review-driven approach, this research offers a promising pathway toward greater digital inclusivity.
As e-commerce continues to grow, ensuring accessibility must remain a top priority. Systems like Revamp set the foundation for a more equitable online shopping experience, empowering VIPs with greater independence and confidence in their purchasing decisions.
Reference
Wang, R., Chen, Z., Zhang, M.R., Li, Z., Liu, Z., Dang, Z., Yu, C. and Chen, X.A., 2021, May. Revamp: Enhancing accessible information seeking experience of online shopping for blind or low vision users. In Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems (pp. 1-14).
Comments