RON: Reverse Connection with Objectness Prior Networks for Object Detection

By Tao Kong, Fuchun Sun, Anbang Yao, Huaping Liu, Ming Lu, Yurong Chen. 2017

419 citations recorded by the source corpus.

We present RON, an efficient and effective framework for generic object detection. Our motivation is to smartly associate the best of the region-based (e.g., Faster R-CNN) and region-free (e.g., SSD) methodologies. Under fully convolutional architecture, RON mainly focuses on two fundamental problems: (a) multi-scale object localization and (b) negative sample mining. To address (a), we design the reverse connection, which enables the network to detect objects on multi-levels of CNNs. To deal with (b), we propose the objectness prior to significantly reduce the searching space of objects. We optimize the reverse connection, objectness prior and object detector jointly by a multi-task loss function, thus RON can directly predict final detection results from all locations of various feature maps. Extensive experiments on the challenging PASCAL VOC 2007, PASCAL VOC 2012 and MS COCO benchmarks demonstrate the competitive performance of RON. Specifically, with VGG-16 and low resolution 384×384 input size, the network gets 81.3% mAP on PASCAL VOC 2007, 80.7% mAP on PASCAL VOC 2012 datasets. Its superiority increases when datasets become larger and more difficult, as demonstrated by the results on the MS COCO dataset. With 1.5G GPU memory at test phase, the speed of the network is 15 FPS, 3 times faster than the Faster R-CNN counterpart. Code will be made publicly available.

Subfields: Computer Vision and Pattern Recognition, Radiology, Nuclear Medicine and Imaging

How researchers see this paper

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

In Computer Vision and Pattern Recognition

In Radiology, Nuclear Medicine and Imaging

Citations in the corpus

Bridging the Gap Between Anchor-Based and Anchor-Free Detection via Adaptive Training Sample Selection (2019, background): “Thereafter, plenty of works are presented to boost its performance in different aspects, such as fusing context information from different layers [24, 12, 69], training from scratch [50, 73], introducing new loss function [33, 6], anchor refinement and matching [66, 67], architecture redesign [21, 2”
NAS-FPN: Learning Scalable Feature Pyramid Architecture for Object Detection (2019, background): “Many recent works [7, 16, 17, 34, 38, 39, 40, 43, 41] propose various crossscale connections or operations to combine features to gen-”
CornerNet: Detecting Objects as Paired Keypoints (2018, background): “In fact, many current detectors (Shrivastava et al., 2016; Fu et al., 2017; Lin et al., 2016; Kong et al., 2017) already adopted networks similar to the hourglass network.”
Single-Shot Refinement Neural Network for Object Detection (2017, methodology): “Current one-stage methods [13, 23, 29] rely on one-step regression based on various feature layers with different scales to predict the locations and sizes of objects, which is rather inaccurate in some challenging scenarios, especially for the small objects.”
CenterNet: Keypoint Triplets for Object Detection (2019, background): “RON [20] proposes a reverse connection and an objectness prior to extract multiscale features effectively.”
Deep Feature Pyramid Reconfiguration for Object Detection (2018, methodology): “More recently, the lateral connection (or reverse connection) is becoming popular and used in object detection [15,30,26].”
Single-Shot Bidirectional Pyramid Networks for High-Quality Object Detection (2018, neutral): “8 One-stage Detectors: RON384 [25] VGG-16 384 × 384 15 75.”
Adaptive Convolution for Object Detection (2019, methodology): “We also take some popular advanced detectors for comparison, including two-stage detectors like Faster R-CNN [3], [11], RFCN [12] and ION [36], one-stage detectors like RON [37].”
Guided Attention Network for Object Detection and Counting on Drones (2019, neutral): “, Faster R-CNN [25], RON [12], SSD [18], RFCN [2], CADNet [5], One-Look Regression [20], IEP [29],”
Triply Supervised Decoder Networks for Joint Detection and Segmentation (2018, methodology): “Among these methods, SSD [31], RON [23], DSSD [10], DES [53], STDN [55], RefineDet [52], and DFPR [22] are only for object detection.”
A survey on instance segmentation: state of the art (2020, methodology): “Many techniques [48,52-54] have been developed for improving the detection accuracy by using various CNN layers.”
W2F: A Weakly-Supervised to Fully-Supervised Framework for Object Detection (2018, neutral): “Recently, many state-of-the-art methods [11, 12, 5, 20] based on deep Convolutional Neural”
Efficient Neural Architecture Transformation Searchin Channel-Level for Object Detection (2019, background): “Many recent works [15, 14, 41] propose various multi-scale integration strategies to generate pyramidal feature representations.”
Imbalance Problems in Object Detection: A Review (2019, methodology): “[56] proposed a method that learns objectness priors in an end-to-end setting in order to have a guidance on where to search for the objects.”
FoveaBox: Beyond Anchor-based Object Detector (2019, background): “Third, since there is a large set of candidate object locations regularly sampled across an image, dense object detectors usually rely on effective techniques to deal with the foreground-background class imbalance challenge [28][23][41].”