A convolutional neural network cascade for face detection

By Haoxiang Li, Zhe L. Lin, Xiaohui Shen, Jonathan Brandt, Gang Hua. 2015

1,329 citations recorded by the source corpus.

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

FairFace: Face Attribute Dataset for Balanced Race, Gender, and Age (2019, background): “…image datasets [22, 31, 13, 69, 37, 24, 41, 68, 15, 27, 46, 8, 39] have been proposed and fostered research and development for automated face detection [35, 21], alignment [66, 44], recognition [57, 49], generation [67, 5, 26, 58], modiﬁcation [3, 32, 19], and attribute classiﬁcation [31, 37].”
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.”
Cascade R-CNN: Delving Into High Quality Object Detection (2017, background): “[21, 28] embedded the classic cascade architecture of [34] in object detection networks.”
Dynamic Neural Networks: A Survey (2021, background): “Image Sa Object detection (face [40], [203], [204], facial point [205], pedestrian [206], general [33], [207], [208], [209], [210]) Image segmentation [107], [211], [212], Super resolution [213], Style transfer [214], Coarse-to-fine classification [215]”
Hybrid Task Cascade for Instance Segmentation (2019, background): “[21] propose to operate at multiple resolutions to reject simple samples.”
Interpreting CNN knowledge via an Explanatory Graph (2017, background): “Convolutional neural networks (CNNs) [12, 10, 9, 14] have achieved superior performance in object classification and detection.”
Spatially Adaptive Computation Time for Residual Networks (2016, methodology): “Cascaded classifiers [27, 43] are used in object detection to quickly reject “easy” negative proposals.”
DenseBox: Unifying Landmark Localization with End to End Object Detection (2015, background): “However, most state-of-the-art object detection approaches [26, 20, 8, 14, 41] rely on R-CNN, which divides detection into two steps: salient object proposal generation and region proposal classiﬁcation.”
Not All Pixels Are Equal: Difficulty-Aware Semantic Segmentation via Deep Layer Cascade (2017, methodology): “[18] used CNN cascade for face detection, which rejects false detections quickly in early stages and carefully reﬁnes detections in later stages.”
Deep Learning for Generic Object Detection: A Survey (2018, background): “xisting work is essential for further progress in object detection, particularly for researchers wishing to enter the ﬁeld. Extensive work on CNNs for speciﬁc object detection, such as face detection [120,237,92], pedestrian detection [238,85], vehicle detection [247] and trafﬁc sign detection [253]”
MCDNN: An Approximation-Based Execution Framework for Deep Stream Processing Under Resource Constraints (2016, background): “Although traditionally detection and tracking have been performed with relatively lightweight algorithms, state-of-the-art variants of these are switching to Deep Neural Networks (DNNs) [29, 33].”
Runtime Neural Pruning (2017, methodology): “Dynamic network: Dynamic network structures and executions have been studied in previous works [7, 28, 33, 39, 40].”
SSH: Single Stage Headless Face Detector (2017, background): “Face detectors based on CNNs significantly closed the performance gap between human and artificial detectors [12, 33, 32, 38, 7].”
Detecting Masked Faces in the Wild with LLE-CNNs (2017, methodology): “[17] proposed CascadeCNN, which is a boosted exemplar-based face detector.”
Energy-efficient Amortized Inference with Cascaded Deep Classifiers (2017, background): “Others [Li et al. , 2015; Yang et al. , 2016] consider cascaded classiﬁers in object detection to quickly reject proposals that are easy to judge.”
Large-Pose Face Alignment via CNN-Based Dense 3D Model Fitting (2016, background): “In [9], a subset of AFLW is selected to have a balanced distribution of yaw angles, including 3, 901 images for training and 1, 299 images for testing.”
S^3FD: Single Shot Scale-Invariant Face Detector (2017, methodology): “On the one hand, many works [21, 31, 54, 55, 58] have applied CNN as the feature extractor in the traditional face detection framewrok.”
Channel Selection Using Gumbel Softmax (2020, background): “Cascaded classifiers [50] shorten computation by identifying easy negatives and have been adapted to deep learning [28, 51].”