DetNet: Design Backbone for Object Detection

By Zeming Li, Chao Peng, Gang Yu, Xiangyu Zhang, Yangdong Deng. 2018

228 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Fast Neural Network Adaptation via Parameter Remapping and Architecture Search (2020, methodology): “DetNet (Li et al., 2018) applies dilated convolution (Yu & Koltun, 2016) in the backbone to enlarge the receptive ﬁeld which helps to detect objects.”
DetNAS: Backbone Search for Object Detection (2019, methodology): “For instance, the recent hand-crafted network, DetNet [12], has demonstrated this point.”
Deep High-Resolution Representation Learning for Visual Recognition (2019, methodology): “The fully convolutional network is extended, by re-placing a few (typically two) strided convolutions and the associated convolutions with dilated convolutions, to the dilation version, leading to medium-resolution representations [18], [19], [86], [168], [181].”
Scale-Aware Trident Networks for Object Detection (2019, methodology): “In the object detection field, DetNet [24] designs a specific detection backbone network to maintain the spatial resolution and enlarge the receptive field using dilated convolution.”
YOLOv4: Optimal Speed and Accuracy of Object Detection (2020, background): “In addition to the above models, some researchers put their emphasis on directly building a new backbone (DetNet [43], DetNAS [7]) or a new whole model (SpineNet [12], HitDetector [20]) for object detection.”
High-Resolution Representations for Labeling Pixels and Regions (2019, background): “k is extended, by replacing a few (typically two) strided convolutions and the associated convolutions with dilated convolutions, to the dilation version, leading to medium-resolution representations [126,13,115,12,57]. The representations are further 1The techniques developed for human pose estimat”
An End-To-End Network for Panoptic Segmentation (2019, methodology): “The proposal-based approaches [8, 14, 24, 25, 28, 29, 33] first generate the object detection bounding boxes and then perform mask prediction on each box for instance segmentation.”
Efficient Training of Visual Transformers with Small Datasets (2021, background): “, in case of 3D point cloud data [65]) or when using a task-specific backbone architecture [32, 34].”
CBNet: A Composite Backbone Network Architecture for Object Detection (2021, background): “, DetNet [44] and FishNet [45], are specifically designed for the detection task.”
SpineNet: Learning Scale-Permuted Backbone for Recognition and Localization (2019, background): “DetNet [20] argues that down-sampling features compromises its localization capability.”
Imbalance Problems in Object Detection: A Review (2019, background): “[106] discuss the cons of employing such networks designed for the classification task and propose a backbone specially designed for the object detection task where they limit the spatial downsampling rate for the high level features.”
You Only Look One-level Feature (2021, methodology): “Although DetNet [18] also successively applies dilated residual blocks, its purpose is to maintain the spatial resolution of the features and keep more details in the backbone’s outputs, while ours is to generate a 1× 1 3× 3 1× 1 3× 3 1× 1”
Efficient Neural Architecture Transformation Searchin Channel-Level for Object Detection (2019, methodology): “For backbones, there are [17, 37] designed specially for object detection manually.”
Dilated SpineNet for Semantic Segmentation (2021, background): “Recent architectures discovered by NAS have shown promising results than handcrafted models on vision tasks including image classification [57, 31, 41, 20], object detection [16, 14, 13, 23, 43, 46, 24], semantic segmentation [27, 34], etc.”
Extended Feature Pyramid Network for Small Object Detection (2020, unknown): “In this section, we ﬁrstly introduce deep learning based general object detectors, and then discuss relevant small object detection methods including utilizing cross-scale features and combining with super-resolution.”