Zoom Out-and-In Network with Map Attention Decision for Region Proposal and Object Detection

By Hongyang Li, Yu Liu, Wanli Ouyang, Xiaogang Wang. 2017

82 citations recorded by the source corpus.

In this paper, we propose a zoom-out-and-in network for generating object proposals. A key observation is that it is difficult to classify anchors of different sizes with the same set of features. Anchors of different sizes should be placed accordingly based on different depth within a network: smaller boxes on high-resolution layers with a smaller stride while larger boxes on low-resolution counterparts with a larger stride. Inspired by the conv/deconv structure, we fully leverage the low-level local details and high-level regional semantics from two feature map streams, which are complimentary to each other, to identify the objectness in an image. A map attention decision (MAD) unit is further proposed to aggressively search for neuron activations among two streams and attend the most contributive ones on the feature learning of the final loss. The unit serves as a decision-maker to adaptively activate maps along certain channels with the solely purpose of optimizing the overall training loss. One advantage of MAD is that the learned weights enforced on each feature channel is predicted on-the-fly based on the input context, which is more suitable than the fixed enforcement of a convolutional kernel. Experimental results on three datasets demonstrate the effectiveness of our proposed algorithm over other state-of-the-arts, in terms of average recall for region proposal and average precision for object detection.

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

Learning Open-World Object Proposals without Learning to Classify (2021, background): “Recently, learning-based proposals [1], [2], [13], [23], [24] have demonstrated better performance than classical approaches in both precision and recall, and are an important part of two-stage detectors.”
Region Proposal by Guided Anchoring (2019, neutral): “M AR L runtime (s/img) SharpMask [27] ResNet-50 36.4 - 48.2 6.0 51.0 66.5 0.76 (unfair) GCN-NS [25] VGG-16 (SyncBN) 31.6 - 60.7 - - - 0.10 AttractioNet [11] VGG-16 53.3 - 66.2 31.5 62.2 77.7 4.00 ZIP [18] BN-inception 53.9 - 67.0 31.9 63.0 78.5 1.13 RPN ResNet-50-FPN 47.5 54.7 59.4 31.7 55.1 64.6 0.”
Neural Network Encapsulation (2018, background): “Convolutional neural networks (CNNs) [1] have been proved to be quite successful in modern deep learning architectures [2,3,4,5] and achieved better performance in various computer vision tasks [6,7,8].”
GS3D: An Efficient 3D Object Detection Framework for Autonomous Driving (2019, background): “Since state-of-the-art 2D detectors [21, 18, 13, 17, 16, 15] can provide reliable 2D bounding boxes for objects,”
Parallel Feature Pyramid Network for Object Detection (2018, background): “It has been shown that the lower- and upper-level features are complementary to each other and their combinations are beneficial for segmentation [30], keypoint estimation [29], and object detection [22].”
Multi-Scale Self-Guided Attention for Medical Image Segmentation (2019, background): “cies. The integration of these attention modules has been proved very successful in many vision problems, such as image captioning [37], image question-answering [38], classiﬁcation [39] or detection [40], among many others. Self-attention [41], [42], [43], [44] has recently attracted the attention ”
Revisiting the Sibling Head in Object Detector (2020, background): “generic object classification task, the specialized sibling head for both classification and localization comes into focus and is widely used in most of advanced detectors including single stage family [25, 33, 12], two-stage family [5, 18, 40, 26, 19] and anchor free family [17].”
xView: Objects in Context in Overhead Imagery (2018, background): “Detecting objects on multiple scales is an ongoing topic of research [19, 3].”
Gradient Harmonized Single-stage Detector (2018, unknown): “(Girshick 2015; Ren et al. 2015; Li et al. 2017; He et al. 2017; Lin et al. 2017a; Zeng et al. 2018).”