Simple Copy-Paste is a Strong Data Augmentation Method for Instance Segmentation

@article{Ghiasi2021SimpleCI,
  title={Simple Copy-Paste is a Strong Data Augmentation Method for Instance Segmentation},
  author={Golnaz Ghiasi and Yin Cui and A. Srinivas and Rui Qian and Tsung-Yi Lin and Ekin Dogus Cubuk and Quoc V. Le and Barret Zoph},
  journal={2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year={2021},
  pages={2917-2927}
}
  • Golnaz Ghiasi, Yin Cui, Barret Zoph
  • Published 13 December 2020
  • Computer Science
  • 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
Building instance segmentation models that are data-efficient and can handle rare object categories is an important challenge in computer vision. Leveraging data augmentations is a promising direction towards addressing this challenge. Here, we perform a systematic study of the Copy-Paste augmentation (e.g., [13], [12]) for instance segmentation where we randomly paste objects onto an image. Prior studies on Copy-Paste relied on modeling the surrounding visual context for pasting the objects… 
Evaluating COPY-BLEND Augmentation for Low Level Vision Tasks
TLDR
This paper examines the efficacy of a simple copy-blend data augmentation technique that copies patches from noisy images and blends onto a clean image and vice versa to ensure that an underlying algorithm localizes and recovers affected regions resulting in increased perceptual quality of a recovered image.
TabAug: Data Driven Augmentation for Enhanced Table Structure Recognition
TLDR
TabAug is proposed, a re-imagined Data Augmentation technique that produces structural changes in table images through replication and deletion of rows and columns through a data-driven probabilistic model that allows control over the augmentation process.
Leveraging background augmentations to encourage semantic focus in self-supervised contrastive learning
TLDR
This work investigates a class of simple, yet highly effective “background augmentations”, which encourage models to focus on semantically-relevant content by discouraging them from focusing on image backgrounds, and demonstrates that background augmentations improve robustness to a number of out of distribution settings.
Crop-Transform-Paste: Self-Supervised Learning for Visual Tracking
TLDR
The Crop-Transform-Paste operation is developed, which is able to synthesize sufficient training data by simulating various kinds of scene variations during tracking, including appearance variations of objects and background changes, and can be combined with supervised learning to further boost the performance.
Object-Based Augmentation Improves Quality of Remote SensingSemantic Segmentation
TLDR
This study proposes a novel pipeline for georeferenced image augmentation that enables a significant increase in the number of training samples and leads to the meaningful improvement of U-Net model predictions from 0.78 to 0.83 F1-score.
Carton dataset synthesis based on foreground texture replacement
TLDR
A novel image synthesis method that can greatly augment the target datasets and improve the model’s performance and a contour reconstruction algorithm is proposed to solve the problem of occlusion, truncation, and incomplete contour of carton instances.
Tackling Catastrophic Forgetting and Background Shift in Continual Semantic Segmentation
TLDR
Local POD is proposed, a multi-scale pooling distillation scheme that preserves long and short-range spatial relationships at feature level and significantly outperforms state-of-the-art methods in existing CSS scenarios, as well as in newly proposed challenging benchmarks.
Data Augmentation Based on Cannikin Law for Object Detection
  • Xue Deng, Hongwei Luo, Hua Zhang
  • Computer Science
    2021 IEEE 6th International Conference on Cloud Computing and Big Data Analytics (ICCCBDA)
  • 2021
TLDR
This paper presents an offline data augmentation method based on Cannikin's Law for augmenting the categories far below the average detection accuracy in multi-class object detection.
Zero-Shot Detection via Vision and Language Knowledge Distillation
TLDR
The goal of this work is to advance zero-shot object detection, which aims to detect novel objects without bounding box nor mask annotations, and proposes ViLD, a training method via Vision and Language knowledge Distillation.
Highly Contrast Image Correction for Dim Boundary Separation of Image Semantic Segmentation
TLDR
The results indicated that the proposed gamma correction could improve image segmentation accuracy in images with unclear edges and will guide further studies for accuracy improvement of image recognition algorithms applicable to future devices, such as autonomous vehicles and mobile robots.
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 78 REFERENCES
InstaBoost: Boosting Instance Segmentation via Probability Map Guided Copy-Pasting
TLDR
This paper presents a simple, efficient and effective method to augment the training set using the existing instance mask annotations, and proposes a location probability map based approach to explore the feasible locations that objects can be placed based on local appearance similarity.
Cut, Paste and Learn: Surprisingly Easy Synthesis for Instance Detection
TLDR
This paper proposes a simple approach to generate large annotated instance datasets with minimal effort and outperforms existing synthesis approaches and when combined with real images improves relative performance by more than 21% on benchmark datasets.
LVIS: A Dataset for Large Vocabulary Instance Segmentation
TLDR
This work introduces LVIS (pronounced ‘el-vis’): a new dataset for Large Vocabulary Instance Segmentation, which has a long tail of categories with few training samples due to the Zipfian distribution of categories in natural images.
Learning to Segment via Cut-and-Paste
TLDR
This paper presents a weakly-supervised approach to object instance segmentation, which exceeds the performance of existing weakly supervised methods, without requiring hand-tuned segment proposals, and reaches 90% of supervised performance.
Microsoft COCO: Common Objects in Context
We present a new dataset with the goal of advancing the state-of-the-art in object recognition by placing the question of object recognition in the context of the broader question of scene
Modeling Visual Context is Key to Augmenting Object Detection Datasets
TLDR
This work leverages segmentation annotations to increase the number of object instances present on training data and shows that modeling appropriately the visual context surrounding objects is crucial to place them in the right environment.
Instance-Aware Semantic Segmentation via Multi-task Network Cascades
  • Jifeng Dai, Kaiming He, Jian Sun
  • Computer Science
    2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
  • 2016
TLDR
This paper presents Multitask Network Cascades for instance-aware semantic segmentation, which consists of three networks, respectively differentiating instances, estimating masks, and categorizing objects, and develops an algorithm for the nontrivial end-to-end training of this causal, cascaded structure.
Simultaneous Detection and Segmentation
TLDR
This work builds on recent work that uses convolutional neural networks to classify category-independent region proposals (R-CNN), introducing a novel architecture tailored for SDS, and uses category-specific, top-down figure-ground predictions to refine the bottom-up proposals.
Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation
TLDR
This paper proposes a simple and scalable detection algorithm that improves mean average precision (mAP) by more than 30% relative to the previous best result on VOC 2012 -- achieving a mAP of 53.3%.
Randaugment: Practical automated data augmentation with a reduced search space
TLDR
This work proposes a simplified search space that vastly reduces the computational expense of automated augmentation, and permits the removal of a separate proxy task.
...
1
2
3
4
5
...