# An Analysis of Pre-Training on Object Detection

arXiv: https://arxiv.org/abs/1904.05871 · HTML: https://mlsanity.com/paper/1904.05871/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 11 Apr 2019.

## Built on by (most cited first, 12 of 12)

- [Multi-Task Self-Training for Learning General Representations](https://mlsanity.com/paper/2108.11353/) (2021)
- [Exploring Visual Interpretability for Contrastive Language-Image Pre-training](https://mlsanity.com/paper/2209.07046/) (2022)
- [GAIA: A Transfer Learning System of Object Detection that Fits Your Needs](https://mlsanity.com/paper/2106.11346/) (2021)
- [Object-Aware Cropping for Self-Supervised Learning](https://mlsanity.com/paper/2112.00319/) (2021)
- [Deep transfer learning for image classification: a survey](https://mlsanity.com/paper/2205.09904/) (2022)
- [Transfer Learning Gaussian Anomaly Detection by Fine-tuning Representations](https://mlsanity.com/paper/2108.04116/) (2021)
- [DAP: Detection-Aware Pre-training with Weak Supervision](https://mlsanity.com/paper/2103.16651/) (2021)
- [A High-Resolution Dataset for Instance Detection with Multi-View Instance Capture](https://mlsanity.com/paper/2310.19257/) (2023)
- [Rethinking Training from Scratch for Object Detection](https://mlsanity.com/paper/2106.03112/) (2021)
- [GPPF: A General Perception Pre-training Framework via Sparsely Activated Multi-Task Learning](https://mlsanity.com/paper/2208.02148/) (2022)
- [Revisiting Pre-training in Audio-Visual Learning](https://mlsanity.com/paper/2302.03533/) (2023)
- [Task adaption by biologically inspired stochastic comodulation](https://mlsanity.com/paper/2311.15053/) (2023)

## Abstract (arXiv)

We provide a detailed analysis of convolutional neural networks which are pre-trained on the task of object detection. To this end, we train detectors on large datasets like OpenImagesV4, ImageNet Localization and COCO. We analyze how well their features generalize to tasks like image classification, semantic segmentation and object detection on small datasets like PASCAL-VOC, Caltech-256, SUN-397, Flowers-102 etc. Some important conclusions from our analysis are --- 1) Pre-training on large detection datasets is crucial for fine-tuning on small detection datasets, especially when precise localization is needed. For example, we obtain 81.1% mAP on the PASCAL-VOC dataset at 0.7 IoU after pre-training on OpenImagesV4, which is 7.6% better than the recently proposed DeformableConvNetsV2 which uses ImageNet pre-training. 2) Detection pre-training also benefits other localization tasks like semantic segmentation but adversely affects image classification. 3) Features for images (like avg. pooled Conv5) which are similar in the object detection feature space are likely to be similar in the image classification feature space but the converse is not true. 4) Visualization of features reveals that detection neurons have activations over an entire object, while activations for classification networks typically focus on parts. Therefore, detection networks are poor at classification when multiple instances are present in an image or when an instance only covers a small fraction of an image.