Exploit All the Layers: Fast and Accurate CNN Object Detector with Scale Dependent Pooling and Cascaded Rejection Classifiers

By Fan Yang, Wongun Choi, Yuanqing Lin. 2016

581 citations recorded by the source corpus.

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

Multi-view 3D Object Detection Network for Autonomous Driving (2016, neutral): “96 SDP+RPN [29, 18] Mono 90.”
Towards Safe Autonomous Driving: Capture Uncertainty in the Deep Neural Network For Lidar 3D Vehicle Detection (2018, background): “Deep learning has been introduced to object detection in the autonomous driving settings that use cameras [1]–[4], Lidar [5]–[9], or both [10]–[15], and has set the benchmark on many popular datasets (e.”
A Unified Multi-scale Deep Convolutional Neural Network for Fast Object Detection (2016, methodology): “We also led a nontrivial margin over the very recent SDP+RPN [42], which used scale depen-”
Spatially Adaptive Computation Time for Residual Networks (2016, methodology): “Cascaded classifiers [27, 43] are used in object detection to quickly reject “easy” negative proposals.”
An Analysis of Scale Invariance in Object Detection - SNIP (2017, methodology): “Methods like SDP [35], SSH [27] or MS-CNN [3], which make independent predictions at different layers, also ensure that smaller objects are trained on higher resolution layers (like conv3) while larger objects are trained on lower resolution layers (like conv5).”
A General Pipeline for 3D Detection of Vehicles (2018, background): “proposed in [21], [7] and [22] can be incorporated since”
Perception, Planning, Control, and Coordination for Autonomous Vehicles (2017, methodology): “To address the scale issue, in [129], Yang et al.”
GSM: Graph Similarity Model for Multi-Object Tracking (2020, methodology): “The detections produced by different object detectors each with increasing performance, namely DPM [Felzenszwalb et al. , 2010], FRCNN [Ren et al. , 2015] and SDP [Yang et al. , 2016].”
NetTailor: Tuning the Architecture, Not Just the Weights (2019, background): “Recent works developed these ideas within a deep learning framework, both for classification [61, 22] and detection [5, 68].”
Deep learning in video multi-object tracking: A survey (2019, methodology): “The MOT17 dataset4 includes the same videos as MOT16, but with more accurate ground truth and with three sets of detections for each video: one from Faster R-CNN [4], one from DPM and one from the Scale-Dependent Pooling detector (SDP) [27].”
Chained-Tracker: Chaining Paired Attentive Regression Results for End-to-End Joint Multiple-Object Detection and Tracking (2020, methodology): “In public detection, MOT16 includes DPM [25] detector while MOT17 includes DPM, Faster R-CNN [4] and SDP [26] detectors.”
Learning a Neural Solver for Multiple Object Tracking (2019, background): “Most state-of-the-art MOT works follow the trackingby-detection paradigm which divides the problem into two steps: (i) detecting pedestrian locations independently in each frame, for which neural networks are currently the state-of-the-art [54, 1, 70], and (ii) linking corresponding detections acros”
Fusing Bird View LIDAR Point Cloud and Front View Camera Image for Deep Object Detection (2017, methodology): “Detecting from vastly different scale is hard for the vanilla Faster-RCNN.”
Optimal Sensor Data Fusion Architecture for Object Detection in Adverse Weather Conditions (2018, background): “[2], [13], [17], where the evaluation is focused on detecting cars, pedestrians and cyclists, the detection accuracy of the proposed approaches is determined for all seven categories.”