Divide and Count: Generic Object Counting by Image Divisions

Stahl, Tobias; Pintea, Silvia L.; Gemert, Jan C. van

doi:10.1109/tip.2018.2875353

Cited by 63 publications

(25 citation statements)

References 48 publications

(119 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Quantitative detection results are provided in Table 3, result curves are presented in Fig. 4, and [22] 22.76 34.46 YOLO9000 opt (2017) [37] 130.40 172.46 RetinaNet (2018) [27] 24.58 33.12 IEP Counting (2019) [41] 15.17 -Our full approach 7.16 12.00 Table 4. CARPK and PUCPR+ counting results [22].…”

Section: Experiments On the Sku-110k Benchmarkmentioning

confidence: 99%

See 1 more Smart Citation

Precise Detection in Densely Packed Scenes

Goldman

Herzig

Eisenschtat³

et al. 2019

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

182

134

View full text Add to dashboard Cite

Man-made scenes can be densely packed, containing numerous objects, often identical, positioned in close proximity. We show that precise object detection in such scenes remains a challenging frontier even for state-of-the-art object detectors. We propose a novel, deep-learning based method for precise object detection, designed for such challenging settings. Our contributions include: (1) A layer for estimating the Jaccard index as a detection quality score;(2) a novel EM merging unit, which uses our quality scores to resolve detection overlap ambiguities; finally, (3) an extensive, annotated data set, SKU-110K, representing packed retail environments, released for training and testing under such extreme settings. Detection tests on SKU-110K and counting tests on the CARPK and PUCPR+ show our method to outperform existing state-of-the-art with substantial margins. The code and data will be made available on

show abstract

Section: Experiments On the Sku-110k Benchmarkmentioning

confidence: 99%

“…Counting results. We compare our method with results reported by others [22,41]: Faster R-CNN [38], YOLO [36], and One-Look Regression [32]. Existing baselines also include two methods designed and tested for counting on these two benchmarks: LPN Counting [22] and IEP Counting [41].…”

Section: Experiments On Carpk and Pucpr+mentioning

confidence: 99%

Precise Detection in Densely Packed Scenes

Goldman

Herzig

Eisenschtat³

et al. 2019

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

182

134

View full text Add to dashboard Cite

show abstract

“…Abstractive text summary using a generative adversarial network was done by the authors in [51], while the authors in [52] proposed a CNN-based technique to obtain high representational features for the detection of secondary protein structures. In order to further improve accuracy, researchers used CNN-based crowd-counting techniques [21,53,54]. Counting through CNN employs convolution, pooling, Rectified Linear Unit (RelU), and Fully Connected Layers (FCLs) to extract features that are used to obtain the density map [55].…”

Section: Counting By Cnnmentioning

confidence: 99%

“…In short, the manual nature of feature extraction by handcrafted methods makes them less (non)adaptive to evolving crowd-counting demands. By observing the above-mentioned deficiencies in traditional crowd-counting algorithms, and the success of CNNs in numerous computer-vision applications, researchers were inspired to exploit their ability in estimating the nonlinear feature density maps of crowd images [53][54][55]. These density maps can be utilized in machine-learning processes for more accurate prediction/estimation of the crowd count [63,64].…”

Section: Motivation For Employing Cnn-based Image Crowd Countingmentioning

confidence: 99%

Convolutional-Neural Network-Based Image Crowd Counting: Review, Categorization, Analysis, and Performance Evaluation

Ilyas

Shahzad

Kim

2019

Sensors

View full text Add to dashboard Cite

Traditional handcrafted crowd-counting techniques in an image are currently transformed via machine-learning and artificial-intelligence techniques into intelligent crowd-counting techniques. This paradigm shift offers many advanced features in terms of adaptive monitoring and the control of dynamic crowd gatherings. Adaptive monitoring, identification/recognition, and the management of diverse crowd gatherings can improve many crowd-management-related tasks in terms of efficiency, capacity, reliability, and safety. Despite many challenges, such as occlusion, clutter, and irregular object distribution and nonuniform object scale, convolutional neural networks are a promising technology for intelligent image crowd counting and analysis. In this article, we review, categorize, analyze (limitations and distinctive features), and provide a detailed performance evaluation of the latest convolutional-neural-network-based crowd-counting techniques. We also highlight the potential applications of convolutional-neural-network-based crowd-counting techniques. Finally, we conclude this article by presenting our key observations, providing strong foundation for future research directions while designing convolutional-neural-network-based crowd-counting techniques. Further, the article discusses new advancements toward understanding crowd counting in smart cities using the Internet of Things (IoT).

show abstract

“…Statistically, small samples are not reliable, but may be the animals having same habitat and living, in that case, a small sample of animals can represent the population. Monitoring the frequency relying on individual animals in the detection and identification process [2] [3]. Estimating the density (or counting) of animals without use of individual recognition may also speed-up the underlying process [10] [26].…”

Section: Introductionmentioning

confidence: 99%

Deep Learning Methods for monitoring, detecting and measuring Deer Movements for Wildlife Conservation

Rawat*,

Sodhi,

Tyagi

2019

IJRTE

View full text Add to dashboard Cite

Wildlife Researchers examine and dig video corpus for behavioral studies of free-ranging animals, which included monitoring, analyzing, classifying & detecting, managing, counting etc. Unfortunately, automated visual implementation for challenging real-time scenarios of wildlife is not an easy task especially for classification and recognition of wildlife-animals and estimate the sizes of wildlife populations. The aim of this paper is to bring state-of-the-art results from raw sensor data for learning features advancing automatic implementation and interpreting of animal movements from different perspectives. Also, turnout with an objectness score from object proposals generated by Region Proposal Network (RPN). The imagery data are captured from the motion sensor cameras and then through RCNN, Fast RCNN and Faster RCNN, it automatically are segmented and recognized the object with its objectness score. ConvNet automatically process these images and correctly recognizing the object. Experimentation results demonstrated prominent deer images with 96% accuracy with identifying three basic activities sleeping, grazing and resting. In addition, a measured implementation has been shown among CNN, RCNN, Fast RCNN and Faster RCNN.

show abstract

Divide and Count: Generic Object Counting by Image Divisions

Cited by 63 publications

References 48 publications

Precise Detection in Densely Packed Scenes

Precise Detection in Densely Packed Scenes

Convolutional-Neural Network-Based Image Crowd Counting: Review, Categorization, Analysis, and Performance Evaluation

Deep Learning Methods for monitoring, detecting and measuring Deer Movements for Wildlife Conservation

Contact Info

Product

Resources

About