# Inverting Visual Representations with Convolutional Networks

arXiv: https://arxiv.org/abs/1506.02753 · HTML: https://mlsanity.com/paper/1506.02753/

## Key facts

- Cited by 170 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 9 Jun 2015.

## Built on by (most cited first, 30 of 170)

- [Parameter-Efficient Fine-Tuning for Large Models: A Comprehensive Survey](https://mlsanity.com/paper/2403.14608/) (2024)
- [Projected GANs Converge Faster](https://mlsanity.com/paper/2111.01007/) (2021)
- [Few-shot Image Generation via Cross-domain Correspondence](https://mlsanity.com/paper/2104.06820/) (2021)
- [Text Embeddings Reveal (Almost) As Much As Text](https://mlsanity.com/paper/2310.06816/) (2023)
- [Squeeze, Recover and Relabel: Dataset Condensation at ImageNet Scale From A New Perspective](https://mlsanity.com/paper/2306.13092/) (2023)
- [Salient ImageNet: How to discover spurious features in Deep Learning?](https://mlsanity.com/paper/2110.04301/) (2021)
- [What do Vision Transformers Learn? A Visual Exploration](https://mlsanity.com/paper/2212.06727/) (2022)
- [Language Model Inversion](https://mlsanity.com/paper/2311.13647/) (2023)
- [Plug & Play Attacks: Towards Robust and Flexible Model Inversion Attacks](https://mlsanity.com/paper/2201.12179/) (2022)
- [Rethinking Reconstruction Autoencoder-Based Out-of-Distribution Detection](https://mlsanity.com/paper/2203.02194/) (2022)
- [Exploiting Explanations for Model Inversion Attacks](https://mlsanity.com/paper/2104.12669/) (2021)
- [Is Geometry Enough for Matching in Visual Localization?](https://mlsanity.com/paper/2203.12979/) (2022)
- [Operationalizing a Threat Model for Red-Teaming Large Language Models (LLMs)](https://mlsanity.com/paper/2407.14937/) (2024)
- [Orchestra: Unsupervised Federated Learning via Globally Consistent Clustering](https://mlsanity.com/paper/2205.11506/) (2022)
- [Extracting Prompts by Inverting LLM Outputs](https://mlsanity.com/paper/2405.15012/) (2024)
- [Logits of API-Protected LLMs Leak Proprietary Information](https://mlsanity.com/paper/2403.09539/) (2024)
- [Biometrics: Trust, but Verify](https://mlsanity.com/paper/2105.06625/) (2021)
- [Defining and Quantifying the Emergence of Sparse Concepts in DNNs](https://mlsanity.com/paper/2111.06206/) (2021)
- [ResSFL: A Resistance Transfer Framework for Defending Model Inversion Attack in Split Federated Learning](https://mlsanity.com/paper/2205.04007/) (2022)
- [A Little Robustness Goes a Long Way: Leveraging Robust Features for Targeted Transfer Attacks](https://mlsanity.com/paper/2106.02105/) (2021)
- [Video Autoencoder: self-supervised disentanglement of static 3D structure and motion](https://mlsanity.com/paper/2110.02951/) (2021)
- [Explaining Deepfake Detection by Analysing Image Matching](https://mlsanity.com/paper/2207.09679/) (2022)
- [Visual Explanations from Spiking Neural Networks using Interspike Intervals](https://mlsanity.com/paper/2103.14441/) (2021)
- [Does a Neural Network Really Encode Symbolic Concepts?](https://mlsanity.com/paper/2302.13080/) (2023)
- [Refining activation downsampling with SoftPool](https://mlsanity.com/paper/2101.00440/) (2021)
- [Transferable Embedding Inversion Attack: Uncovering Privacy Risks in Text Embeddings without Model Queries](https://mlsanity.com/paper/2406.10280/) (2024)
- [Privacy-preserving Adversarial Facial Features](https://mlsanity.com/paper/2305.05391/) (2023)
- [Privacy-Preserving Face Recognition Using Trainable Feature Subtraction](https://mlsanity.com/paper/2403.12457/) (2024)
- [NinjaDesc: Content-Concealing Visual Descriptors via Adversarial Learning](https://mlsanity.com/paper/2112.12785/) (2021)
- [A Survey of What to Share in Federated Learning: Perspectives on Model Utility, Privacy Leakage, and Communication Efficiency](https://mlsanity.com/paper/2307.10655/) (2023)

## Abstract (arXiv)

Feature representations, both hand-designed and learned ones, are often hard to analyze and interpret, even when they are extracted from visual data. We propose a new approach to study image representations by inverting them with an up-convolutional neural network. We apply the method to shallow representations (HOG, SIFT, LBP), as well as to deep networks. For shallow representations our approach provides significantly better reconstructions than existing methods, revealing that there is surprisingly rich information contained in these features. Inverting a deep network trained on ImageNet provides several insights into the properties of the feature representation learned by the network. Most strikingly, the colors and the rough contours of an image can be reconstructed from activations in higher network layers and even from the predicted class probabilities.