# Multiple Instance Curriculum Learning for Weakly Supervised Object Detection

arXiv: https://arxiv.org/abs/1711.09191 · HTML: https://mlsanity.com/paper/1711.09191/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 25 Nov 2017.

## Built on by (most cited first, 13 of 13)

- [Curriculum Learning: A Survey](https://mlsanity.com/paper/2101.10382/) (2021)
- [Weakly Supervised Object Localization and Detection: A Survey](https://mlsanity.com/paper/2104.07918/) (2021)
- [STPro: Spatial and Temporal Progressive Learning for Weakly Supervised Spatio-Temporal Grounding](https://mlsanity.com/paper/2502.20678/) (2025)
- [RecRecNet: Rectangling Rectified Wide-Angle Images by Thin-Plate Spline Model and DoF-based Curriculum Learning](https://mlsanity.com/paper/2301.01661/) (2023)
- [Counterfactual Cross-modality Reasoning for Weakly Supervised Video Moment Localization](https://mlsanity.com/paper/2308.05648/) (2023)
- [Approaching Outside: Scaling Unsupervised 3D Object Detection from 2D Scene](https://mlsanity.com/paper/2407.08569/) (2024)
- [Learning 3D Perception from Others' Predictions](https://mlsanity.com/paper/2410.02646/) (2024)
- [Accelerating the Training of Video Super-Resolution Models](https://mlsanity.com/paper/2205.05069/) (2022)
- [Are Dense Labels Always Necessary for 3D Object Detection from Point Cloud?](https://mlsanity.com/paper/2403.02818/) (2024)
- [Curriculum for Crowd Counting -- Is it Worthy?](https://mlsanity.com/paper/2401.07586/) (2024)
- [False Positive Sampling-based Data Augmentation for Enhanced 3D Object Detection Accuracy](https://mlsanity.com/paper/2403.02639/) (2024)
- [Learning to Anticipate Future with Dynamic Context Removal](https://mlsanity.com/paper/2204.02587/) (2022)
- [Online progressive instance-balanced sampling for weakly supervised object detection](https://mlsanity.com/paper/2206.10324/) (2022)

## Abstract (arXiv)

When supervising an object detector with weakly labeled data, most existing approaches are prone to trapping in the discriminative object parts, e.g., finding the face of a cat instead of the full body, due to lacking the supervision on the extent of full objects. To address this challenge, we incorporate object segmentation into the detector training, which guides the model to correctly localize the full objects. We propose the multiple instance curriculum learning (MICL) method, which injects curriculum learning (CL) into the multiple instance learning (MIL) framework. The MICL method starts by automatically picking the easy training examples, where the extent of the segmentation masks agree with detection bounding boxes. The training set is gradually expanded to include harder examples to train strong detectors that handle complex images. The proposed MICL method with segmentation in the loop outperforms the state-of-the-art weakly supervised object detectors by a substantial margin on the PASCAL VOC datasets.