Pedestrian detection aided by deep learning semantic tasks

By Yonglong Tian, Ping Luo, Xiaogang Wang, Xiaoou Tang. Computer Vision and Pattern Recognition, 2014

440 citations recorded by the source corpus.

Deep learning methods have achieved great successes in pedestrian detection, owing to its ability to learn discriminative features from raw pixels. However, they treat pedestrian detection as a single binary classification task, which may confuse positive with hard negative samples (Fig.1 (a)). To address this ambiguity, this work jointly optimize pedestrian detection with semantic tasks, including pedestrian attributes (e.g. `carrying backpack') and scene attributes (e.g. `vehicle', `tree', and `horizontal'). Rather than expensively annotating scene attributes, we transfer attributes information from existing scene segmentation datasets to the pedestrian dataset, by proposing a novel deep model to learn high-level features from multiple tasks and multiple data sources. Since distinct tasks have distinct convergence rates and data from different datasets have different distributions, a multi-task deep model is carefully designed to coordinate tasks and reduce discrepancies among datasets. Extensive evaluations show that the proposed approach outperforms the state-of-the-art on the challenging Caltech [9] and ETH [10] datasets where it reduces the miss rates of previous deep models by 17 and 5.5 percent, respectively.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/21d2d64408fcfbe6f2c8d0cb53a614dcdb16ede5

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

Deep Learning Strong Parts for Pedestrian Detection (2015, methodology): “[32] modeled detection task together with attribute prediction tasks within a single deep model.”
Multi-Task Convolutional Neural Network for Pose-Invariant Face Recognition (2017, methodology): “It has been successfully applied to face detection [7], [60], face alignment [63], pedestrian detection [51],”
Filtered channel features for pedestrian detection (2015, background): “) Other features include bag-of-words over colour, HOG, or LBP features [4]; learning sparse dictionary encoders [33]; and training features via a convolutional neural network [35] ( [37, 16] appeared while preparing this manuscript).”
Taking a deeper look at pedestrians (2015, background): “While preparing this manuscript, additional related papers appeared [46, 54].”
Illumination-aware Faster R-CNN for Robust Multispectral Pedestrian Detection (2018, background): “Most of initial attempts [5, 6, 17, 18, 19] to apply convnets for pedestrian detection task adopted a two-stage pipeline in an R-CNN [20] style.”
Object Detection in 20 Years: A Survey (2019, neutral): “pedestrians) [316].”
Perceptual Generative Adversarial Networks for Small Object Detection (2017, methodology): “We compare the result of Perceptual GAN with all the existing methods that achieved best performance on the Caltech testing set, including VJ [37], HOG [5], LDCF [27], Katamari [2], SpatialPooling+ [30], TA-CNN [36], Checkerboards [43], CompACT-Deep [44] !" #$%%# &'(”
Object Detection With Deep Learning: A Review (2018, methodology): “[206] proposed a taskassistant CNN (TA-CNN) to jointly learn multiple tasks with TABLE VII DETAILED BREAKDOWN PERFORMANCE COMPARISONS OF STATE-OF-THE-ART MODELS ON CALTECH PEDESTRIAN DATASET.”
Deep Learning Face Attributes in the Wild (2014, background): “Deep learning [18, 34, 23, 7, 19, 32, 31, 13, 33, 22, 3, 28] recently achieved great success in attribute prediction, due to their ability to learn compact and discriminative features.”
Data-Free Knowledge Amalgamation via Group-Stack Dual-GAN (2020, background): “Multi-task learning(MTL) has been widely studied in many fields, including computer vision [8, 1], natural language processing [4], and machine learning [26, 41].”
Bi-box Regression for Pedestrian Detection and Occlusion Estimation (2018, methodology): “A deep CNN is learned to jointly optimize pedestrian detection and other semantic tasks to improve pedestrian detection performance [32].”
Learning Efficient Single-Stage Pedestrian Detectors by Asymptotic Localization Fitting (2018, methodology): “In TA-CNN [38], the ACF detector [10] is employed for proposal generation, then pedestrian detection is jointly optimized with an auxiliary semantic task.”
Scale-Aware Fast R-CNN for Pedestrian Detection (2015, methodology): “Some recent works focus on improving the performance of pedestrian detection using deep learning methods [24], [32], [23], [35].”
Is Faster R-CNN Doing Well for Pedestrian Detection? (2016, methodology): “Despite the prevalent success of deeply learned features in computer vision, current leading pedestrian detectors ( e.g. , [3,4,5,6]) are in general hybrid methods that combines traditional, hand-crafted features [7,8] and deep convolutional features [9,10].”
Multispectral Deep Neural Networks for Pedestrian Detection (2016, methodology): “[40] improved pedestrian detection by learning high-level features from DNNs of multiple tasks, including pedestrian attribute prediction.”
A Brief Review of Facial Emotion Recognition Based on Visual Information (2018, neutral): “First, LSTM models are straightforward in terms”