# Deep Feature Augmentation for Occluded Image Classification

arXiv: https://arxiv.org/abs/2011.00768 · HTML: https://mlsanity.com/paper/2011.00768/

## Key facts

- Cited by 21 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 2 Nov 2020.

## Built on by (most cited first, 21 of 21)

- [A Comprehensive Survey of Image Augmentation Techniques for Deep Learning](https://mlsanity.com/paper/2205.01491/) (2022)
- [Watch or Listen: Robust Audio-Visual Speech Recognition with Visual Corruption Modeling and Reliability Scoring](https://mlsanity.com/paper/2303.08536/) (2023)
- [Colonoscopy Polyp Detection and Classification: Dataset Creation and Comparative Evaluations](https://mlsanity.com/paper/2104.10824/) (2021)
- [Enhanced U-Net: A Feature Enhancement Network for Polyp Segmentation](https://mlsanity.com/paper/2105.00999/) (2021)
- [SOSD-Net: Joint Semantic Object Segmentation and Depth Estimation from Monocular images](https://mlsanity.com/paper/2101.07422/) (2021)
- [Accumulated Trivial Attention Matters in Vision Transformers on Small Datasets](https://mlsanity.com/paper/2210.12333/) (2022)
- [SGNet: A Super-class Guided Network for Image Classification and Object Detection](https://mlsanity.com/paper/2104.12898/) (2021)
- [Audio-Visual Transformer Based Crowd Counting](https://mlsanity.com/paper/2109.01926/) (2021)
- [Few-Shot Learning by Integrating Spatial and Frequency Representation](https://mlsanity.com/paper/2105.05348/) (2021)
- [An Unsupervised Domain Adaptation Model based on Dual-module Adversarial Training](https://mlsanity.com/paper/2112.15555/) (2021)
- [Semantic Clustering based Deduction Learning for Image Recognition and Classification](https://mlsanity.com/paper/2112.13165/) (2021)
- [Dynamic Label Assignment for Object Detection by Combining Predicted IoUs and Anchor IoUs](https://mlsanity.com/paper/2201.09396/) (2022)
- [Are Deep Learning Models Robust to Partial Object Occlusion in Visual Recognition Tasks?](https://mlsanity.com/paper/2409.10775/) (2024)
- [A Discriminative Channel Diversification Network for Image Classification](https://mlsanity.com/paper/2112.05861/) (2021)
- [UpCycling: Semi-supervised 3D Object Detection without Sharing Raw-level Unlabeled Scenes](https://mlsanity.com/paper/2211.11950/) (2022)
- [Multiple Classifiers Based Maximum Classifier Discrepancy for Unsupervised Domain Adaptation](https://mlsanity.com/paper/2108.00610/) (2021)
- [Towards More Effective PRM-based Crowd Counting via A Multi-resolution Fusion and Attention Network](https://mlsanity.com/paper/2112.09664/) (2021)
- [Explicitly Increasing Input Information Density for Vision Transformers on Small Datasets](https://mlsanity.com/paper/2210.14319/) (2022)
- [Latent Enhancing AutoEncoder for Occluded Image Classification](https://mlsanity.com/paper/2402.06936/) (2024)
- [Training Deep Neural Networks via Branch-and-Bound](https://mlsanity.com/paper/2104.01730/) (2021)
- [The JEPA Predictor: A Transferable Operator for Occluded Feature Completion](https://mlsanity.com/paper/2607.16274/) (2026)

## Abstract (arXiv)

Due to the difficulty in acquiring massive task-specific occluded images, the classification of occluded images with deep convolutional neural networks (CNNs) remains highly challenging. To alleviate the dependency on large-scale occluded image datasets, we propose a novel approach to improve the classification accuracy of occluded images by fine-tuning the pre-trained models with a set of augmented deep feature vectors (DFVs). The set of augmented DFVs is composed of original DFVs and pseudo-DFVs. The pseudo-DFVs are generated by randomly adding difference vectors (DVs), extracted from a small set of clean and occluded image pairs, to the real DFVs. In the fine-tuning, the back-propagation is conducted on the DFV data flow to update the network parameters. The experiments on various datasets and network structures show that the deep feature augmentation significantly improves the classification accuracy of occluded images without a noticeable influence on the performance of clean images. Specifically, on the ILSVRC2012 dataset with synthetic occluded images, the proposed approach achieves 11.21% and 9.14% average increases in classification accuracy for the ResNet50 networks fine-tuned on the occlusion-exclusive and occlusion-inclusive training sets, respectively.