Abstract
Many modern object detectors demonstrate outstanding performances by using the mechanism of looking and thinking twice. In this paper, we explore this mechanism in the backbone design for object detection. At the macro level, we propose Recursive Feature Pyramid, which incorporates extra feedback connections from Feature Pyramid Networks into the bottom-up backbone layers. At the micro level, we propose Switchable Atrous Convolution, which convolves the features with different atrous rates and gathers the results using switch functions. Combining them results in DetectoRS, which significantly improves the performances of object detection. On COCO test-dev, DetectoRS achieves state-of-the-art 55.7% box AP for object detection, 48.5% mask AP for instance segmentation, and 50.0% PQ for panoptic segmentation. The code is made publicly available <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sup> .
Keywords
Pyramid (geometry)Convolution (computer science)DetectorObject detectionComputer scienceFeature (linguistics)SegmentationArtificial intelligenceCode (set theory)Object (grammar)Feature extractionPattern recognition (psychology)Computer visionMathematicsArtificial neural networkProgramming language
Affiliated Institutions
Related Publications
EfficientDet: Scalable and Efficient Object Detection
Model efficiency has become increasingly important in computer vision. In this paper, we systematically study neural network architecture design choices for object detection and...
Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions
Although convolutional neural networks (CNNs) have achieved great success in computer vision, this work investigates a simpler, convolution-free backbone network use-fid for man...
CSPNet: A New Backbone that can Enhance Learning Capability of CNN
Neural networks have enabled state-of-the-art approaches to achieve incredible results on computer vision tasks such as object detection. However, such success greatly relies on...
FCOS: Fully Convolutional One-Stage Object Detection
We propose a fully convolutional one-stage object detector (FCOS) to solve object detection in a per-pixel prediction fashion, analogue to semantic segmentation. Almost all stat...
Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation
Spatial pyramid pooling module or encode-decoder structure are used in deep neural networks for semantic segmentation task. The former networks are able to encode multi-scale co...
Publication Info
- Year
- 2021
- Type
- article
- Pages
- 10208-10219
- Citations
- 942
- Access
- Closed
External Links
Social Impact
Altmetric
PlumX Metrics
Social media, news, blog, policy document mentions
Citation Metrics
942
OpenAlex
Cite This
Siyuan Qiao,
Liang-Chieh Chen,
Alan Yuille
(2021).
DetectoRS: Detecting Objects with Recursive Feature Pyramid and Switchable Atrous Convolution.
, 10208-10219.
https://doi.org/10.1109/cvpr46437.2021.01008
Identifiers
- DOI
- 10.1109/cvpr46437.2021.01008