# Bounding boxes for weakly supervised segmentation: Global constraints get close to full supervision

arXiv: https://arxiv.org/abs/2004.06816 · HTML: https://mlsanity.com/paper/2004.06816/

## Key facts

- Cited by 40 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 14 Apr 2020.

## Built on by (most cited first, 30 of 40)

- [Source-Free Domain Adaptation for Image Segmentation](https://mlsanity.com/paper/2108.03152/) (2021)
- [Every Annotation Counts: Multi-label Deep Supervision for Medical Image Segmentation](https://mlsanity.com/paper/2104.13243/) (2021)
- [Semantic Segmentation In-the-Wild Without Seeing Any Segmentation Examples](https://mlsanity.com/paper/2112.03185/) (2021)
- [FedMix: Mixed Supervised Federated Learning for Medical Image Segmentation](https://mlsanity.com/paper/2205.01840/) (2022)
- [Medical Image Segmentation with Limited Supervision: A Review of Deep Network Models](https://mlsanity.com/paper/2103.00429/) (2021)
- [Inter Extreme Points Geodesics for End-to-End Weakly Supervised Image Segmentation](https://mlsanity.com/paper/2107.00583/) (2021)
- [Bounding Box Tightness Prior for Weakly Supervised Image Segmentation](https://mlsanity.com/paper/2110.00934/) (2021)
- [Weakly supervised segmentation with cross-modality equivariant constraints](https://mlsanity.com/paper/2104.02488/) (2021)
- [PA-Seg: Learning from Point Annotations for 3D Medical Image Segmentation using Contextual Regularization and Cross Knowledge Distillation](https://mlsanity.com/paper/2208.05669/) (2022)
- [Prototype-Based Image Prompting for Weakly Supervised Histopathological Image Segmentation](https://mlsanity.com/paper/2503.12068/) (2025)
- [The surprising impact of mask-head architecture on novel class segmentation](https://mlsanity.com/paper/2104.00613/) (2021)
- [Greybox XAI: a Neural-Symbolic learning framework to produce interpretable predictions for image classification](https://mlsanity.com/paper/2209.14974/) (2022)
- [Versatile Medical Image Segmentation Learned from Multi-Source Datasets via Model Self-Disambiguation](https://mlsanity.com/paper/2311.10696/) (2023)
- [Weakly-Supervised 3D Medical Image Segmentation using Geometric Prior and Contrastive Similarity](https://mlsanity.com/paper/2302.02125/) (2023)
- [Negative Evidence Matters in Interpretable Histology Image Classification](https://mlsanity.com/paper/2201.02445/) (2022)
- [Segmentation with mixed supervision: Confidence maximization helps knowledge distillation](https://mlsanity.com/paper/2109.10902/) (2021)
- [Weakly Supervised Intracranial Hemorrhage Segmentation using Head-Wise Gradient-Infused Self-Attention Maps from a Swin Transformer in Categorical Learning](https://mlsanity.com/paper/2304.04902/) (2023)
- [Industrial Anomaly Detection and Localization Using Weakly-Supervised Residual Transformers](https://mlsanity.com/paper/2306.03492/) (2023)
- [FedLPPA: Learning Personalized Prompt and Aggregation for Federated Weakly-supervised Medical Image Segmentation](https://mlsanity.com/paper/2402.17502/) (2024)
- [LNQ 2023 challenge: Benchmark of weakly-supervised techniques for mediastinal lymph node quantification](https://mlsanity.com/paper/2408.10069/) (2024)
- [Automating MedSAM by Learning Prompts with Weak Few-Shot Supervision](https://mlsanity.com/paper/2409.20293/) (2024)
- [Self-supervised Multi-scale Consistency for Weakly Supervised Segmentation Learning](https://mlsanity.com/paper/2108.11900/) (2021)
- [A comprehensive survey on recent deep learning-based methods applied to surgical data](https://mlsanity.com/paper/2209.01435/) (2022)
- [Weakly Supervised Image Segmentation Beyond Tight Bounding Box Annotations](https://mlsanity.com/paper/2301.12053/) (2023)
- [Weakly Supervised Volumetric Segmentation via Self-taught Shape Denoising Model](https://mlsanity.com/paper/2104.13082/) (2021)
- [CDRNet: Accurate Cup-to-Disc Ratio Measurement with Tight Bounding Box Supervision in Fundus Photography Using Deep Learning](https://mlsanity.com/paper/2110.00943/) (2021)
- [Leveraging point annotations in segmentation learning with boundary loss](https://mlsanity.com/paper/2311.03537/) (2023)
- [Prompt learning with bounding box constraints for medical image segmentation](https://mlsanity.com/paper/2507.02743/) (2025)
- [Robust Trust Region for Weakly Supervised Segmentation](https://mlsanity.com/paper/2104.01948/) (2021)
- [Medical image segmentation with imperfect 3D bounding boxes](https://mlsanity.com/paper/2108.03300/) (2021)

## Abstract (arXiv)

We propose a novel weakly supervised learning segmentation based on several global constraints derived from box annotations. Particularly, we leverage a classical tightness prior to a deep learning setting via imposing a set of constraints on the network outputs. Such a powerful topological prior prevents solutions from excessive shrinking by enforcing any horizontal or vertical line within the bounding box to contain, at least, one pixel of the foreground region. Furthermore, we integrate our deep tightness prior with a global background emptiness constraint, guiding training with information outside the bounding box. We demonstrate experimentally that such a global constraint is much more powerful than standard cross-entropy for the background class. Our optimization problem is challenging as it takes the form of a large set of inequality constraints on the outputs of deep networks. We solve it with sequence of unconstrained losses based on a recent powerful extension of the log-barrier method, which is well-known in the context of interior-point methods. This accommodates standard stochastic gradient descent (SGD) for training deep networks, while avoiding computationally expensive and unstable Lagrangian dual steps and projections. Extensive experiments over two different public data sets and applications (prostate and brain lesions) demonstrate that the synergy between our global tightness and emptiness priors yield very competitive performances, approaching full supervision and outperforming significantly DeepCut. Furthermore, our approach removes the need for computationally expensive proposal generation. Our code is shared anonymously.