# Return of the Devil in the Details: Delving Deep into Convolutional Nets

arXiv: https://arxiv.org/abs/1405.3531 · HTML: https://mlsanity.com/paper/1405.3531/

## Key facts

- Cited by 122 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 14 May 2014.

## Built on by (most cited first, 30 of 122)

- [Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time](https://mlsanity.com/paper/2203.05482/) (2022)
- [Masked Feature Prediction for Self-Supervised Visual Pre-Training](https://mlsanity.com/paper/2112.09133/) (2021)
- [Probing the 3D Awareness of Visual Foundation Models](https://mlsanity.com/paper/2404.08636/) (2024)
- [On-Device Training Under 256KB Memory](https://mlsanity.com/paper/2206.15472/) (2022)
- [Re-labeling ImageNet: from Single to Multi-Labels, from Global to Localized Labels](https://mlsanity.com/paper/2101.05022/) (2021)
- [Learning Accurate Dense Correspondences and When to Trust Them](https://mlsanity.com/paper/2101.01710/) (2021)
- [The Expressive Power of Low-Rank Adaptation](https://mlsanity.com/paper/2310.17513/) (2023)
- [Vision-Language Models in Remote Sensing: Current Progress and Future Trends](https://mlsanity.com/paper/2305.05726/) (2023)
- [Self-Supervised Video Forensics by Audio-Visual Anomaly Detection](https://mlsanity.com/paper/2301.01767/) (2023)
- [PDC-Net+: Enhanced Probabilistic Dense Correspondence Network](https://mlsanity.com/paper/2109.13912/) (2021)
- [A Broad Study on the Transferability of Visual Representations with Contrastive Learning](https://mlsanity.com/paper/2103.13517/) (2021)
- [Deep Learning for Instance Retrieval: A Survey](https://mlsanity.com/paper/2101.11282/) (2021)
- [DreamMatcher: Appearance Matching Self-Attention for Semantically-Consistent Text-to-Image Personalization](https://mlsanity.com/paper/2402.09812/) (2024)
- [Learning by Aligning Videos in Time](https://mlsanity.com/paper/2103.17260/) (2021)
- [Visual Object Tracking with Discriminative Filters and Siamese Networks: A Survey and Outlook](https://mlsanity.com/paper/2112.02838/) (2021)
- [Voice-Face Homogeneity Tells Deepfake](https://mlsanity.com/paper/2203.02195/) (2022)
- [One Loss for All: Deep Hashing with a Single Cosine Similarity based Learning Objective](https://mlsanity.com/paper/2109.14449/) (2021)
- [Warp Consistency for Unsupervised Learning of Dense Correspondences](https://mlsanity.com/paper/2104.03308/) (2021)
- [Tiny Machine Learning: Progress and Futures](https://mlsanity.com/paper/2403.19076/) (2024)
- [Is a Caption Worth a Thousand Images? A Controlled Study for Representation Learning](https://mlsanity.com/paper/2207.07635/) (2022)
- [Benchmarking Omni-Vision Representation through the Lens of Visual Realms](https://mlsanity.com/paper/2207.07106/) (2022)
- [Does progress on ImageNet transfer to real-world datasets?](https://mlsanity.com/paper/2301.04644/) (2023)
- [Enable Deep Learning on Mobile Devices: Methods, Systems, and Applications](https://mlsanity.com/paper/2204.11786/) (2022)
- [Perspectives on individual animal identification from biology and computer vision](https://mlsanity.com/paper/2103.00560/) (2021)
- [Learning Long-Term Spatial-Temporal Graphs for Active Speaker Detection](https://mlsanity.com/paper/2207.07783/) (2022)
- [PockEngine: Sparse and Efficient Fine-tuning in a Pocket](https://mlsanity.com/paper/2310.17752/) (2023)
- [A Comprehensive Survey of Deep Transfer Learning for Anomaly Detection in Industrial Time Series: Methods, Applications, and Directions](https://mlsanity.com/paper/2307.05638/) (2023)
- [Diffusion Model for Dense Matching](https://mlsanity.com/paper/2305.19094/) (2023)
- [Learning to Align Sequential Actions in the Wild](https://mlsanity.com/paper/2111.09301/) (2021)
- [Just Add π! Pose Induced Video Transformers for Understanding Activities of Daily Living](https://mlsanity.com/paper/2311.18840/) (2023)

## Abstract (arXiv)

The latest generation of Convolutional Neural Networks (CNN) have achieved impressive results in challenging benchmarks on image recognition and object detection, significantly raising the interest of the community in these methods. Nevertheless, it is still unclear how different CNN methods compare with each other and with previous state-of-the-art shallow representations such as the Bag-of-Visual-Words and the Improved Fisher Vector. This paper conducts a rigorous evaluation of these new techniques, exploring different deep architectures and comparing them on a common ground, identifying and disclosing important implementation details. We identify several useful properties of CNN-based representations, including the fact that the dimensionality of the CNN output layer can be reduced significantly without having an adverse effect on performance. We also identify aspects of deep and shallow methods that can be successfully shared. In particular, we show that the data augmentation techniques commonly applied to CNN-based methods can also be applied to shallow methods, and result in an analogous performance boost. Source code and models to reproduce the experiments in the paper is made publicly available.