DSSD : Deconvolutional Single Shot Detector

By Cheng-Yang Fu, Wei David Liu, Ananth Ranga, Ambrish Tyagi, Alexander C. Berg. 2017

2,065 citations recorded by the source corpus.

The main contribution of this paper is an approach for introducing additional context into state-of-the-art general object detection. To achieve this we first combine a state-of-the-art classifier (Residual-101[14]) with a fast detection framework (SSD[18]). We then augment SSD+Residual-101 with deconvolution layers to introduce additional large-scale context in object detection and improve accuracy, especially for small objects, calling our resulting system DSSD for deconvolutional single shot detector. While these two contributions are easily described at a high-level, a naive implementation does not succeed. Instead we show that carefully adding additional stages of learned transformations, specifically a module for feed-forward connections in deconvolution and a new output module, enables this new approach and forms a potential way forward for further detection research. Results are shown on both PASCAL VOC and COCO detection. Our DSSD with $513 \times 513$ input achieves 81.5% mAP on VOC2007 test, 80.0% mAP on VOC2012 test, and 33.2% mAP on COCO, outperforming a state-of-the-art method R-FCN[3] on each dataset.

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

Bridging the Gap Between Anchor-Based and Anchor-Free Detection via Adaptive Training Sample Selection (2019, background): “…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,…”
Speed/Accuracy Trade-Offs for Modern Convolutional Object Detectors (2016, methodology): “[32, 28, 29, 9] use SSD-like architectures to predict final (1 of K) class labels.”
Towards Real-World Prohibited Item Detection: A Large-Scale X-ray Benchmark (2021, methodology): “After that, different one-stage object detectors are proposed, including SSD [24], DSSD [8], and YOLO series [28, 29, 30].”
Focal Loss for Dense Object Detection (2017, background): “More recently SSD [9], [10] and YOLO [7], [8] have renewed interest in one-stage methods.”
Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation (2018, methodology): “Encoder-decoder: The encoder-decoder networks have been successfully applied to many computer vision tasks, including human pose estimation [53], object detection [45, 66, 19], and semantic segmentation [49, 54, 61, 3, 43, 59, 57, 33, 76, 20].”
Path Aggregation Network for Instance Segmentation (2018, methodology): “Different from TDM, which took the fused feature map with the highest resolution to pool features, SSD [42], DSSD [13], MS-CNN [5] and FPN [35] assigned proposals to appropriate feature levels for inference.”
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-”
StairNet: Top-Down Semantic Aggregation for Accurate One Shot Detection (2017, methodology): “Moreover, the recent studies [8,26] have shown that the sequence of deconvolution layer is suitable for propagating the information efficiently.”
CornerNet: Detecting Objects as Paired Keypoints (2018, methodology): “Anchor boxes are extensively used in one-stage detectors (Liu et al., 2016; Fu et al., 2017; Redmon and Farhadi, 2016; Lin et al., 2017), which can achieve results highly competitive with two-stage detectors (Ren et al., 2015; Girshick et al., 2014; Girshick, 2015; He et al., 2017) while being more…”
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, methodology): “DSSD [9] introduces a deconvolution module into SSD to combine low- and high-level features.”
RODEO: Replay for Online Object Detection (2020, neutral): “Although RODEO uses Fast-RCNN, a two stage detector, which is slower than single stage detectors like SSD [13, 34] and YOLO [45, 46], single stage approaches could be used to facilitate faster learning and inference.”
Scale-Aware Trident Networks for Object Detection (2019, background): “Based on the multi-layer prediction module in SSD, DSSD [14] introduces additional context information with deconvolutional operators to improve the accuracy.”
Bottom-Up Object Detection by Grouping Extreme and Center Points (2019, background): “SSD [10,28] uses different scale anchors in different network layers.”
Feature Selective Anchor-Free Module for Single-Shot Object Detection (2019, methodology): “FPN [21] and DSSD [8] proposed to enhance low-level features with high-level semantic feature maps at all scales.”
Modeling Visual Context is Key to Augmenting Object Detection Datasets (2018, methodology): “1% mAP on VOC07-test when trained on the union of the full training and validation parts of VOC’07 and VOC’12, namely VOC07-train+val and VOC12train+val (see [23]); this network is similar to the DSSD detector of [24] that was also used in the Focal Loss paper [37].”