Learning Region Features for Object Detection

By Jiayuan Gu, Han Hu, Liwei Wang, Yichen Wei, Jifeng Dai. 2018

82 citations recorded by the source corpus.

While most steps in the modern object detection methods are learnable, the region feature extraction step remains largely hand-crafted, featured by RoI pooling methods. This work proposes a general viewpoint that unifies existing region feature extraction methods and a novel method that is end-to-end learnable. The proposed method removes most heuristic choices and outperforms its RoI pooling counterparts. It moves further towards fully learnable object detection.

Subfields: Aerospace Engineering, 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 Aerospace Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

Bridging the Gap Between Anchor-Based and Anchor-Free Detection via Adaptive Training Sample Selection (2019, methodology): “As shown in Table 4, different values of k in [3, 5, 7, 9, 11, 13, 15, 17, 19] are used to train the detector.”
Swin Transformer: Hierarchical Vision Transformer using Shifted Windows (2021, background): “The selfattention layers can complement backbones [61, 7, 3, 65, 21, 68, 51] or head networks [29, 24] by providing the capability to encode distant dependencies or heterogeneous interactions.”
Deformable ConvNets V2: More Deformable, Better Results (2018, positive): “In [18], a learnable region feature extractor is proposed, unifying the previous region feature extraction modules from the pixel-object relation perspective.”
An Empirical Study of Spatial Attention Mechanisms in Deep Networks (2019, methodology): “The success of attention modeling in NLP has led to its adoption in computer vision, where different variants of Transformer attention are applied to recognition tasks such as object detection and semantic segmentation [20, 38, 17, 22, 46, 13], where the query and key are visual elements such as ima”
Group-Free 3D Object Detection via Transformers (2021, methodology): “…a sampling method to generate initial object candidates, and stacked attention modules to extract and reﬁne object representations from all points. image recognition to this paper are those who apply the attention mechanism or Transformer architectures into 2D object detection [15, 11, 5, 2].”
Recent Advances in Deep Learning for Object Detection (2019, background): “[139] proposed a fully learnable object detector which proposed a general viewpoint that unified existing region feature extraction methods.”
RelationNet++: Bridging Visual Representations for Object Detection via Transformer Decoder (2020, methodology): “Learning Region Features (LRF) [10] and DeTr [1] use an attention module to compute the features of object proposals [10] or querys [1] from image features.”
A Comparative Analysis of Object Detection Metrics with a Companion Open-Source Toolkit (2021, methodology): “In this context, graph neural networks (GNNs) are employed to compute region of interest features in a more efficient way and process the objects simultaneously by modeling them according to their appearance feature and geometry [55,56].”
Disentangled Non-Local Neural Networks (2020, background): “Since then, such modeling methods have demonstrated great effectiveness in many vision tasks, such as image classification [22], object detection [21,15], semantic segmentation [39], video object detection [10,36,16,6] and tracking [37], and action recognition [34].”
Counterfactual Attention Learning for Fine-Grained Visual Categorization and Re-identification (2021, unknown): “Many efforts [57, 65, 50, 18, 45, 14, 1, 6, 36] have been made to model the mechanism of human attention in computer vision systems, which aim to facilitate high-performance recognition by discovering discriminative regions and mitigating the negative effects brought by diverse visual appearance, cl”