# RetinaMask: Learning to predict masks improves state-of-the-art single-shot detection for free

arXiv: https://arxiv.org/abs/1901.03353 · HTML: https://mlsanity.com/paper/1901.03353/

## Key facts

- Cited by 24 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 10 Jan 2019.

## Built on by (most cited first, 24 of 24)

- [OpenForensics: Large-Scale Challenging Dataset For Multi-Face Forgery Detection And Segmentation In-The-Wild](https://mlsanity.com/paper/2107.14480/) (2021)
- [SOTR: Segmenting Objects with Transformers](https://mlsanity.com/paper/2108.06747/) (2021)
- [Multi-branch Collaborative Learning Network for 3D Visual Grounding](https://mlsanity.com/paper/2407.05363/) (2024)
- [Camouflaged Instance Segmentation In-The-Wild: Dataset, Method, and Benchmark Suite](https://mlsanity.com/paper/2103.17123/) (2021)
- [A Finite Element-Inspired Hypergraph Neural Network: Application to Fluid Dynamics Simulations](https://mlsanity.com/paper/2212.14545/) (2022)
- [RF-DETR Object Detection vs YOLOv12 : A Study of Transformer-based and CNN-based Architectures for Single-Class and Multi-Class Greenfruit Detection in Complex Orchard Environments Under Label Ambiguity](https://mlsanity.com/paper/2504.13099/) (2025)
- [VideoClick: Video Object Segmentation with a Single Click](https://mlsanity.com/paper/2101.06545/) (2021)
- [EMAG: Ego-motion Aware and Generalizable 2D Hand Forecasting from Egocentric Videos](https://mlsanity.com/paper/2405.20030/) (2024)
- [Human-centric Relation Segmentation: Dataset and Solution](https://mlsanity.com/paper/2105.11168/) (2021)
- [CenterPoly: real-time instance segmentation using bounding polygons](https://mlsanity.com/paper/2108.08923/) (2021)
- [Mask-aware IoU for Anchor Assignment in Real-time Instance Segmentation](https://mlsanity.com/paper/2110.09734/) (2021)
- [DarkVision: A Benchmark for Low-light Image/Video Perception](https://mlsanity.com/paper/2301.06269/) (2023)
- [COMICS: End-to-end Bi-grained Contrastive Learning for Multi-face Forgery Detection](https://mlsanity.com/paper/2308.01520/) (2023)
- [Mask or Non-Mask? Robust Face Mask Detector via Triplet-Consistency Representation Learning](https://mlsanity.com/paper/2110.00523/) (2021)
- [Robust Deepfake On Unrestricted Media: Generation And Detection](https://mlsanity.com/paper/2202.06228/) (2022)
- [CenDerNet: Center and Curvature Representations for Render-and-Compare 6D Pose Estimation](https://mlsanity.com/paper/2208.09829/) (2022)
- [Contour Loss for Instance Segmentation via k-step Distance Transformation Image](https://mlsanity.com/paper/2102.10854/) (2021)
- [SODAR: Segmenting Objects by DynamicallyAggregating Neighboring Mask Representations](https://mlsanity.com/paper/2202.07402/) (2022)
- [EffiPerception: an Efficient Framework for Various Perception Tasks](https://mlsanity.com/paper/2403.12317/) (2024)
- [Analysis of Object Detection Models for Tiny Object in Satellite Imagery: A Dataset-Centric Approach](https://mlsanity.com/paper/2412.10453/) (2024)
- [Boundary Distribution Estimation for Precise Object Detection](https://mlsanity.com/paper/2111.01396/) (2021)
- [Generalized Mask-aware IoU for Anchor Assignment for Real-time Instance Segmentation](https://mlsanity.com/paper/2312.17031/) (2023)
- [On depth prediction for autonomous driving using self-supervised learning](https://mlsanity.com/paper/2403.06194/) (2024)
- [DiffClean: Diffusion-based Makeup Removal for Accurate Age Estimation](https://mlsanity.com/paper/2507.13292/) (2025)

## Abstract (arXiv)

Recently two-stage detectors have surged ahead of single-shot detectors in the accuracy-vs-speed trade-off. Nevertheless single-shot detectors are immensely popular in embedded vision applications. This paper brings single-shot detectors up to the same level as current two-stage techniques. We do this by improving training for the state-of-the-art single-shot detector, RetinaNet, in three ways: integrating instance mask prediction for the first time, making the loss function adaptive and more stable, and including additional hard examples in training. We call the resulting augmented network RetinaMask. The detection component of RetinaMask has the same computational cost as the original RetinaNet, but is more accurate. COCO test-dev results are up to 41.4 mAP for RetinaMask-101 vs 39.1mAP for RetinaNet-101, while the runtime is the same during evaluation. Adding Group Normalization increases the performance of RetinaMask-101 to 41.7 mAP. Code is at:https://github.com/chengyangfu/retinamask