Image-Based Product Listing With Concurrent Attribute Recognition

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Users often lack detailed knowledge about second-hand products, leading to inaccurate information filling during publication on second-hand product websites, resulting in poor searchability and reduced selling success.

Innovation Solution

An image-based product object publishing method that utilizes an image recognizer set to automatically recognize and provide structured product attribute information, allowing users to select and publish the product accurately.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If users manually fill in product information on second-hand product websites, then the publishing process can be completed, but the accuracy of product information is poor because users do not know the product details well

Engineering Contradiction:

Improveaccuracy of product informationVSAvoiddifficulty of filling product information

Core Design Contradiction:

Measurement precisionVSEase of operation

Solution Approach 1:

The system enables the product object itself to provide information through image recognition. The image recognizer automatically extracts product attributes (brand, model, category, etc.) from the uploaded product image, allowing the product to 'self-describe' without requiring user knowledge of product specifications.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The manual mechanical process of users typing and selecting product information is replaced by an automated image recognition system. The recognizer set processes the product image and automatically generates structured product information, substituting human manual input with automated optical and computational processes.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Productivity

If users manually search for and select product attributes from dropdown menus, then structured information can be obtained, but the publishing efficiency is low and time-consuming

Engineering Contradiction:

Improvepublishing efficiencyVSAvoidtime for filling product information

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The system performs preliminary action by automatically recognizing and extracting product attributes from the image before the user needs to publish. The image recognizer set processes the image and generates structured product information in advance, so when the user submits the publication, the information is already prepared and structured.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The product image itself serves as the source of information. The recognition system extracts attributes directly from visual features of the product, eliminating the need for users to manually search through dropdown menus or type information, thereby dramatically improving publishing efficiency.

Inventive Principle:

Principle #25Self-service

3Loss of information

If the system provides detailed product attribute fields for users to fill, then complete product information can be collected, but users are overwhelmed and cannot fill them accurately

Engineering Contradiction:

Improvecompleteness of product informationVSAvoidcomplexity of publishing form

Core Design Contradiction:

Loss of informationVSDevice complexity

Solution Approach 1:

The system extracts the essential product information directly from the product image using the image recognizer set. Instead of presenting users with a complex form containing many fields, the system automatically extracts key attributes (brand, model, category, specifications) from the image and pre-fills them, reducing the form to only essential user inputs.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The complex manual process of navigating through multiple dropdown menus and filling numerous fields is replaced by automated image analysis. The recognition system processes the image and generates structured information, transforming a complex multi-step manual process into a simple automated operation.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS12541782B2Method, system, and computer-readable storage medium for product object publishing and concurrent image recognition

Publication Date: 2026.02.03 ALIBABA GROUP HOLDING LTD

AI summary

Embodiments of the specification provide a product object publishing method and a system. The method includes: obtaining an image of a product object; sending the image to a server; receiving a plurality of types of product attribute information obtained by image recognition on the image of the product object performed by an image recognizer set; displaying the plurality of types of product attribute information for selection by a user; generating structured information of the product object according to selection of the product attribute information; and publishing the product object, wherein the publishing comprises publishing the image and the structured information of the product object. Accuracy and efficiency of product object publishing can be improved.