# Exploiting Web Images for Dataset Construction: A Domain Robust Approach

arXiv: https://arxiv.org/abs/1611.07156 · HTML: https://mlsanity.com/paper/1611.07156/

## Key facts

- Cited by 22 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 22 Nov 2016.

## Built on by (most cited first, 22 of 22)

- [Non-Salient Region Object Mining for Weakly Supervised Semantic Segmentation](https://mlsanity.com/paper/2103.14581/) (2021)
- [Jo-SRC: A Contrastive Approach for Combating Noisy Labels](https://mlsanity.com/paper/2103.13029/) (2021)
- [Hierarchical Feature Alignment Network for Unsupervised Video Object Segmentation](https://mlsanity.com/paper/2207.08485/) (2022)
- [Webly Supervised Fine-Grained Recognition: Benchmark Datasets and An Approach](https://mlsanity.com/paper/2108.02399/) (2021)
- [FECANet: Boosting Few-Shot Semantic Segmentation with Feature-Enhanced Context-Aware Network](https://mlsanity.com/paper/2301.08160/) (2023)
- [Spatial Structure Constraints for Weakly Supervised Semantic Segmentation](https://mlsanity.com/paper/2401.11122/) (2024)
- [Semantically Meaningful Class Prototype Learning for One-Shot Image Semantic Segmentation](https://mlsanity.com/paper/2102.10935/) (2021)
- [Trustable Co-label Learning from Multiple Noisy Annotators](https://mlsanity.com/paper/2203.04199/) (2022)
- [Co-attention Propagation Network for Zero-Shot Video Object Segmentation](https://mlsanity.com/paper/2304.03910/) (2023)
- [Manifold DivideMix: A Semi-Supervised Contrastive Learning Framework for Severe Label Noise](https://mlsanity.com/paper/2308.06861/) (2023)
- [DatasetAgent: A Novel Multi-Agent System for Auto-Constructing Datasets from Real-World Images](https://mlsanity.com/paper/2507.08648/) (2025)
- [DiscrimLoss: A Universal Loss for Hard Samples and Incorrect Samples Discrimination](https://mlsanity.com/paper/2208.09884/) (2022)
- [Semi-Supervised Semantic Segmentation With Region Relevance](https://mlsanity.com/paper/2304.11539/) (2023)
- [CAPro: Webly Supervised Learning with Cross-Modality Aligned Prototypes](https://mlsanity.com/paper/2310.09761/) (2023)
- [Relating CNN-Transformer Fusion Network for Change Detection](https://mlsanity.com/paper/2407.03178/) (2024)
- [PanorAMS: Automatic Annotation for Detecting Objects in Urban Context](https://mlsanity.com/paper/2208.14295/) (2022)
- [FoPro: Few-Shot Guided Robust Webly-Supervised Prototypical Learning](https://mlsanity.com/paper/2212.00465/) (2022)
- [Exploiting Web Images for Fine-Grained Visual Recognition by Eliminating Noisy Samples and Utilizing Hard Ones](https://mlsanity.com/paper/2101.09412/) (2021)
- [Attention Map Guided Transformer Pruning for Edge Device](https://mlsanity.com/paper/2304.01452/) (2023)
- [Group Benefits Instances Selection for Data Purification](https://mlsanity.com/paper/2403.15694/) (2024)
- [A Light-weight Transformer-based Self-supervised Matching Network for Heterogeneous Images](https://mlsanity.com/paper/2404.19311/) (2024)
- [Anti-Collapse Loss for Deep Metric Learning Based on Coding Rate Metric](https://mlsanity.com/paper/2407.03106/) (2024)

## Abstract (arXiv)

Labelled image datasets have played a critical role in high-level image understanding. However, the process of manual labelling is both time-consuming and labor intensive. To reduce the cost of manual labelling, there has been increased research interest in automatically constructing image datasets by exploiting web images. Datasets constructed by existing methods tend to have a weak domain adaptation ability, which is known as the "dataset bias problem". To address this issue, we present a novel image dataset construction framework that can be generalized well to unseen target domains. Specifically, the given queries are first expanded by searching the Google Books Ngrams Corpus to obtain a rich semantic description, from which the visually non-salient and less relevant expansions are filtered out. By treating each selected expansion as a "bag" and the retrieved images as "instances", image selection can be formulated as a multi-instance learning problem with constrained positive bags. We propose to solve the employed problems by the cutting-plane and concave-convex procedure (CCCP) algorithm. By using this approach, images from different distributions can be kept while noisy images are filtered out. To verify the effectiveness of our proposed approach, we build an image dataset with 20 categories. Extensive experiments on image classification, cross-dataset generalization, diversity comparison and object detection demonstrate the domain robustness of our dataset.