Chained Cascade Network for Object Detection

By Wanli Ouyang, Kun Wang, Xin Zhu, Xiaogang Wang. 2017

56 citations recorded by the source corpus.

Cascade is a widely used approach that rejects obvious negative samples at early stages for learning better classifier and faster inference. This paper presents chained cascade network (CC-Net). In this CC-Net, there are many cascade stages. Preceding cascade stages are placed at shallow layers. Easy hard examples are rejected at shallow layers so that the computation for deeper or wider layers is not required. In this way, features and classifiers at latter stages handle more difficult samples with the help of features and classifiers in previous stages. It yields consistent boost in detection performance on PASCAL VOC 2007 and ImageNet for both fast RCNN and Faster RCNN. CC-Net saves computation for both training and testing. Code is available on https://github.com/wk910930/ccnn.

Subfields: 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 Computer Vision and Pattern Recognition

Citations in the corpus

Cascade R-CNN: High Quality Object Detection and Instance Segmentation (2019, background): “On the other hand, [33], [43] embedded the classic cascade architecture of [53] in an object detection network.”
Hybrid Task Cascade for Instance Segmentation (2019, background): “For example, CC-Net [31] rejects easy RoIs at shallow layers.”
Libra R-CNN: Towards Balanced Learning for Object Detection (2019, methodology): “Other methods focus on cascade procedures [24], duplicate removal [14, 13], multi-scales [2, 1, 31, 30], adversarial learning [37] and more contextual [36].”
Object Detection in 20 Years: A Survey (2019, methodology): “tc. In addition to the algorithm acceleration, cascaded detection has been applied to solve other problems, e.g., to improve the detection of hard examples [182–184], to integrate context information [143, 185], and to improve localization accuracy [104, 125]. 3.4 Network Pruning and Quantiﬁcation “”
PreDet: Large-scale weakly supervised pre-training for detection (2021, background): “Over the years, various models [30, 38, 42, 20, 33, 55, 65, 66, 52] have been proposed for object detection and instance segmentation.”
Accurate Monocular 3D Object Detection via Color-Embedded 3D Reconstruction for Autonomous Driving (2019, background): “In recent years, with the development of technologies in computer vision and deep learning [12, 34, 36], numerous impressive methods are proposed for accurate 2D object detection [9, 8, 11, 32, 17, 24, 41, 18].”
Deep Learning for Generic Object Detection: A Survey (2018, background): “ond stage classifying the remaining regions. Recently, end to end learning of more than two cascaded classiﬁers and DCNNs for generic object detection was ﬁrst proposed in the Chained Cascade Network [201] and then extended in the Cascade RCNN [23]. Recently, cascade is further applied for simultane”
Quantization Mimic: Towards Very Tiny CNN for Object Detection (2018, background): “The target of object detection [5,6,7,8,9,10] is to locate and classify the objects in images.”
Multi-Adversarial Faster-RCNN for Unrestricted Object Detection (2019, background): “Benefited by the success of CNN models [14, 17, 34], a number of CNN based object detection methods [1, 8, 20, 24, 33, 39] have been emerged.”
Optical Flow Guided Feature: A Fast and Robust Motion Representation for Video Action Recognition (2017, background): “Intermediate supervision has been proven to be practical training strategy in many other computer vision tasks [22, 45, 46, 24, 6].”