# FastMask: Segment Multi-scale Object Candidates in One Shot

arXiv: https://arxiv.org/abs/1612.08843 · HTML: https://mlsanity.com/paper/1612.08843/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 28 Dec 2016.

## Built on by (most cited first, 10 of 10)

- [A Guide to Image and Video based Small Object Detection using Deep Learning : Case Study of Maritime Surveillance](https://mlsanity.com/paper/2207.12926/) (2022)
- [Vision Transformers Are Good Mask Auto-Labelers](https://mlsanity.com/paper/2301.03992/) (2023)
- [Localizing Small Apples in Complex Apple Orchard Environments](https://mlsanity.com/paper/2202.11372/) (2022)
- [What is Point Supervision Worth in Video Instance Segmentation?](https://mlsanity.com/paper/2404.01990/) (2024)
- [Superpixel-based Refinement for Object Proposal Generation](https://mlsanity.com/paper/2101.04574/) (2021)
- [DeepFH Segmentations for Superpixel-based Object Proposal Refinement](https://mlsanity.com/paper/2108.03503/) (2021)
- [SOS: Segment Object System for Open-World Instance Segmentation With Object Priors](https://mlsanity.com/paper/2409.14627/) (2024)
- [Efficient Concertormer for Image Deblurring and Beyond](https://mlsanity.com/paper/2404.06135/) (2024)
- [Small, but important: Traffic light proposals for detecting small traffic lights and beyond](https://mlsanity.com/paper/2307.15191/) (2023)
- [Segmenting Medical Instruments in Minimally Invasive Surgeries using AttentionMask](https://mlsanity.com/paper/2203.11358/) (2022)

## Abstract (arXiv)

Objects appear to scale differently in natural images. This fact requires methods dealing with object-centric tasks (e.g. object proposal) to have robust performance over variances in object scales. In the paper, we present a novel segment proposal framework, namely FastMask, which takes advantage of hierarchical features in deep convolutional neural networks to segment multi-scale objects in one shot. Innovatively, we adapt segment proposal network into three different functional components (body, neck and head). We further propose a weight-shared residual neck module as well as a scale-tolerant attentional head module for efficient one-shot inference. On MS COCO benchmark, the proposed FastMask outperforms all state-of-the-art segment proposal methods in average recall being 2~5 times faster. Moreover, with a slight trade-off in accuracy, FastMask can segment objects in near real time (~13 fps) with 800*600 resolution images, demonstrating its potential in practical applications. Our implementation is available on https://github.com/voidrank/FastMask.