This dataset is collected from Google’s Open Images dataset, using Tencent ML-Images annotations. 4 somewhat similar classes were chosen to create a non-trivial classification problem: mobile phones, laptops, cameras and TVs.
Wu, Baoyuan and Chen, Weidong and Fan, Yanbo and Zhang, Yong and Hou, Jinlong and Liu, Jie and Zhang, Tong
Tencent ML-Images: A Large-Scale Multi-Label Image Database for Visual Representation Learning, IEEE Access, 7, 2019.
The dataset is available under the Creative Commons Attribution International (CC-BY 4.0). Note that the license includes a disclaimer of warranties and limitation of liability.