Domain-adaptive Data Synthesis for Large-scale Supermarket Product Recognition
<p><strong>Domain-Adaptive Data Synthesis for Large-Scale Supermarket Product Recognition</strong></p>
<p>This repository contains the data synthesis pipeline and synthetic product recognition datasets proposed in [1].</p>
<p><strong>Data Synthesis Pipeline:</strong></p>
<p>We provide the Blender 3.1 project files and Python source code of our data synthesis pipeline <em>pipeline.zip, </em>accompanied by the<em> </em><a href="https://github.com/taesungp/contrastive-unpaired-translation">FastCUT</a> models used for synthetic-to-real domain translation<em> models.zip</em>. For the synthesis of new shelf images, a product assortment list and product images must be provided in the corresponding directories <em>products/assortment/</em> and <em>products/img/</em>. The pipeline expects product images to follow the naming convention <em>c</em>.png, with <em>c</em> corresponding to a GTIN or generic class label (e.g., 9120050882171.png). The assortment list, <em>assortment.csv</em>, is expected to use the sample format [<em>c, w, d, h</em>], with <em>c</em> being the class label and <em>w, d,</em> and <em>h</em> being the packaging dimensions of the given product in mm (e.g., [4004218143128, 140, 70, 160]). The assortment list to use and the number of images to generate can be specified in <em>generateImages.py </em>(see comments). The rendering process is initiated by either executing <em>load.py</em> from within Blender or within a command-line terminal as a background process. </p>
<p><strong>Datasets:</strong></p>
<ul>
<li><strong>SG3k</strong> - Synthetic GroZi-3.2k (SG3k) dataset, consisting of 10,000 synthetic shelf images with 851,801 instances of 3,234 GroZi-3.2k products. Instance-level bounding boxes and generic class labels are provided for all product instances.</li>
<li><strong>SG3kt</strong> - Domain-translated version of SGI3k, utilizing GroZi-3.2k as the target domain. Instance-level bounding boxes and generic class labels are provided for all product instances.</li>
<li><strong>SGI3k</strong> - Synthetic GroZi-3.2k (SG3k) dataset, consisting of 10,000 synthetic shelf images with 838,696 instances of 1,063 GroZi-3.2k products. Instance-level bounding boxes and generic class labels are provided for all product instances.</li>
<li><strong>SGI3kt</strong> - Domain-translated version of SGI3k, utilizing GroZi-3.2k as the target domain. Instance-level bounding boxes and generic class labels are provided for all product instances.</li>
<li><strong>SPS8k</strong> - Synthetic Product Shelves 8k (SPS8k) dataset, comprised of 16,224 synthetic shelf images with 1,981,967 instances of 8,112 supermarket products. Instance-level bounding boxes and GTIN class labels are provided for all product instances.</li>
<li><strong>SPS8kt</strong> - Domain-translated version of SPS8k, utilizing SKU110k as the target domain. Instance-level bounding boxes and GTIN class labels for all product instances.</li>
</ul>
<p>Table 1: Dataset characteristics. </p>
<table>
<tbody>
<tr>
<td><strong>Dataset</strong></td>
<td><strong>#images</strong></td>
<td><strong>#products</strong></td>
<td><strong>#instances</strong></td>
<td> <strong>labels </strong></td>
<td><strong>translation</strong></td>
</tr>
<tr>
<td>SG3k</td>
<td>10,000</td>
<td>3,234</td>
<td>851,801</td>
<td>bounding box & generic class¹</td>
<td>none</td>
</tr>
<tr>
<td>SG3kt</td>
<td>10,000</td>
<td>3,234</td>
<td>851,801</td>
<td>bounding box & generic class¹</td>
<td>GroZi-3.2k</td>
</tr>
<tr>
<td>SGI3k</td>
<td>10,000</td>
<td>1,063</td>
<td>838,696</td>
<td>bounding box & generic class²</td>
<td>none</td>
</tr>
<tr>
<td>SGI3kt</td>
<td>10,000</td>
<td>1,063</td>
<td>838,696</td>
<td>bounding box & generic class²</td>
<td>GroZi-3.2k</td>
</tr>
<tr>
<td>SPS8k</td>
<td>16,224</td>
<td>8,112</td>
<td>1,981,967</td>
<td>bounding box & GTIN</td>
<td>none</td>
</tr>
<tr>
<td>SPS8kt</td>
<td>16,224</td>
<td>8,112</td>
<td>1,981,967</td>
<td>bounding box & GTIN</td>
<td>SKU110k</td>
</tr>
</tbody>
</table>
<p> </p>
<p><strong>Sample Format</strong></p>
<p>A sample consists of an RGB image (i.png) and an accompanying label file (i.txt), which contains the labels for all product instances present in the image. Labels use the YOLO format [c, x, y, w, h].</p>
<p>¹SG3k and SG3kt use generic pseudo-GTIN class labels, created by combining the GroZi-3.2k food product category number <em>i</em> (1-27) with the product image index <em>j </em>(j.jpg)<em>, </em>following the convention<em> i0000j </em>(e.g., 13000097).</p>
<p>²SGI3k and SGI3kt use the generic GroZi-3.2k class labels from <a href="https://arxiv.org/abs/2003.06800">https://arxiv.org/abs/2003.06800</a>.</p>
<p><strong>Download and Use</strong><br>This data may be used for non-commercial research purposes only. If you publish material based on this data, we request that you include a reference to our paper [1].</p>
<p>[1] Strohmayer, Julian, and Martin Kampel. "Domain-Adaptive Data Synthesis for Large-Scale Supermarket Product Recognition." <em>International Conference on Computer Analysis of Images and Patterns</em>. Cham: Springer Nature Switzerland, 2023.</p>
<p>BibTeX citation:</p>
<pre>@inproceedings{strohmayer2023domain,
title={Domain-Adaptive Data Synthesis for Large-Scale Supermarket Product Recognition},
author={Strohmayer, Julian and Kampel, Martin},
booktitle={International Conference on Computer Analysis of Images and Patterns},
pages={239--250},
year={2023},
organization={Springer}
}</pre>
opencc-by-4.0Sep 2023View details →
