DeNet: Scalable Real-time Object Detection with Directed Sparse Sampling

Tychsen-Smith, Lachlan

doi:10.48550/arxiv.1703.10295

Cited by 3 publications

(4 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Faster RCNN [41] further replaces region proposals [47] with a Region Proposal Network. The detection-by-classification idea is intuitive and keeps the best performance so far [6,7,19,20,24,27,37,45,46,54].…”

Section: Related Workmentioning

confidence: 99%

Bottom-Up Object Detection by Grouping Extreme and Center Points

Zhou

Zhuo

Krähenbühl

2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

942

477

View full text Add to dashboard Cite

With the advent of deep learning, object detection drifted from a bottom-up to a top-down recognition problem. State of the art algorithms enumerate a near-exhaustive list of object locations and classify each into: object or not. In this paper, we show that bottom-up approaches still perform competitively. We detect four extreme points (top-most, leftmost, bottom-most, right-most) and one center point of objects using a standard keypoint estimation network. We group the five keypoints into a bounding box if they are geometrically aligned. Object detection is then a purely appearance-based keypoint estimation problem, without region classification or implicit feature learning. The proposed method performs on-par with the state-of-the-art region based detection methods, with a bounding box AP of 43.7% on COCO test-dev. In addition, our estimated extreme points directly span a coarse octagonal mask, with a COCO Mask AP of 18.9%, much better than the Mask AP of vanilla bounding boxes. Extreme point guided segmentation further improves this to 34.6% Mask AP.

show abstract

Section: Related Workmentioning

confidence: 99%

Bottom-Up Object Detection by Grouping Extreme and Center Points

Zhou

Zhuo

Krähenbühl

2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

942

477

View full text Add to dashboard Cite

show abstract

“…SSD [22] further improves performance by producing predictions of different scales from different layers. Unlike box-center based detectors, DeNet [36] first predicts all boxes corners, and then quickly searching the corner distribution for non-trivial bounding boxes.…”

Section: Related Workmentioning

confidence: 99%

Light-Head R-CNN: In Defense of Two-Stage Object Detector

Li¹,

Peng²,

Yu³

et al. 2017

Preprint

View full text Add to dashboard Cite

In this paper, we first investigate why typical two-stage methods are not as fast as single-stage, fast detectors like YOLO [26,27] and SSD [22]. We find that Faster R-CNN [28] and R-FCN [17] perform an intensive computation after or before RoI warping. Faster R-CNN involves two fully connected layers for RoI recognition, while R-FCN produces a large score maps. Thus, the speed of these networks is slow due to the heavy-head design in the architecture. Even if we significantly reduce the base model, the computation cost cannot be largely decreased accordingly.We propose a new two-stage detector, Light-Head R-CNN, to address the shortcoming in current two-stage approaches. In our design, we make the head of network as light as possible, by using a thin feature map and a cheap R-CNN subnet (pooling and single fully-connected layer). Our ResNet-101 based light-head R-CNN outperforms state-of-art object detectors on COCO while keeping time efficiency. More importantly, simply replacing the backbone with a tiny network (e.g, Xception), our Light-Head R-CNN gets 30.7 mmAP at 102 FPS on COCO, significantly outperforming the single-stage, fast detectors like YOLO [26,27] and SSD [22] on both speed and accuracy. Code will be made publicly available.

show abstract

“…DeNet [16] relies on region-of-interest estimation based on corner detection. DeNet predicts for each image pixel how likely it is a certain corner of an object bounding box.…”

Section: A Base Detectorsmentioning

confidence: 99%

“…You Only Look Once [12], Single-Shot Detector [11] introduced more lightweight approach and end-to-end training achieving remarkable accuracy while operating in real-time. YOLOv2 [13] and DSSD [4] and most recent DeNet [16] object detector push these boundaries even further.…”

mentioning

confidence: 99%

ALFA: Agglomerative Late Fusion Algorithm for Object Detection

Razinkov

Савельева

Matas

2018

2018 24th International Conference on Pattern Recognition (ICPR)

View full text Add to dashboard Cite

We propose ALFA -a novel late fusion algorithm for object detection. ALFA is based on agglomerative clustering of object detector predictions taking into consideration both the bounding box locations and the class scores. Each cluster represents a single object hypothesis whose location is a weighted combination of the clustered bounding boxes.ALFA was evaluated using combinations of a pair (SSD and DeNet) and a triplet (SSD, DeNet and Faster R-CNN) of recent object detectors that are close to the state-of-the-art. ALFA achieves state of the art results on PASCAL VOC 2007 and PASCAL VOC 2012, outperforming the individual detectors as well as baseline combination strategies, achieving up to 32% lower error than the best individual detectors and up to 6% lower error than the reference fusion algorithm DBF -Dynamic Belief Fusion.

show abstract

DeNet: Scalable Real-time Object Detection with Directed Sparse Sampling

Cited by 3 publications

References 0 publications

Bottom-Up Object Detection by Grouping Extreme and Center Points

Bottom-Up Object Detection by Grouping Extreme and Center Points

Light-Head R-CNN: In Defense of Two-Stage Object Detector

ALFA: Agglomerative Late Fusion Algorithm for Object Detection

Contact Info

Product

Resources

About