DeepFlux for Skeletons in the Wild

Wang, Yukang; Xu, Yongchao; Tsogkas, Stavros; Bai, Xiang; Dickinson, Sven; Siddiqi, Kaleem

doi:10.1109/cvpr.2019.00543

Cited by 44 publications

(20 citation statements)

References 49 publications

(124 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In another convincing dataset SK506 [5], GFF also embodies superiority. GFF achieves the F-score of 72.3%, which is higher than Hi-Fi [8] and DeepFlux [37] with margins of 4.2% and 2.3% separately.…”

Section: Performance and Comparisonmentioning

confidence: 89%

See 1 more Smart Citation

Genetic Feature Fusion for Object Skeleton Detection

Qiao

Tian

Liu

et al. 2021

Security and Communication Networks

View full text Add to dashboard Cite

Object skeleton detection requires the convolutional neural networks to recognize objects and their parts in the cluttered background, overcome the image definition degradation brought by the pooling layers, and predict the location of skeleton pixels in different scale granularity. Most existing object skeleton detection methods take great efforts into the designing of side-output networks for multiscale feature fusion. Despite the great progress achieved by them, there are still many problems that hinder the development of object skeleton detection, such as the manually designed network is labor-intensive and the network initialization depends on models pretrained on large-scale datasets. To alleviate these issues, we propose a genetic NAS method to automatically search on a newly designed architecture search space for adaptive multiscale feature fusion. Furthermore, we introduce a symmetric encoder-decoder search space based on reversing the VGG network, in which the decoder can reuse the ImageNet pretrained model of VGG. The searched networks improve the performance of the state-of-the-art methods on commonly used skeleton detection benchmarks, which proves the efficacy of our method.

show abstract

Section: Performance and Comparisonmentioning

confidence: 89%

“…On the SK-LARGE [30] dataset, our GFF achieves the performance of F-score 73.6%, which is the highest object skeleton detection performance compared to the optimal method DeepFlux [37], which achieves 73.2%. We also outperform Hi-Fi [8], which utilizes the additional scaleassociated ground truth with a margin of 1.2%.…”

Section: Performance and Comparisonmentioning

confidence: 95%

Genetic Feature Fusion for Object Skeleton Detection

Qiao

Tian

Liu

et al. 2021

Security and Communication Networks

View full text Add to dashboard Cite

show abstract

“…In this paper, we propose a novel text detector deemed TextField for detecting texts of arbitrary shapes and orientations. Inspired by component tree representation [16], [50]- [52] that links neighboring pixels following their intensity order to form candidate characters, we propose to learn a deep direction field, which is similar to the notion of flux image [53], to link neighboring pixels and form candidate text parts. The learned direction information is further used to group text parts into text instances.…”

Section: Introductionmentioning

confidence: 99%

TextField: Learning a Deep Direction Field for Irregular Scene Text Detection

Wang

Zhou

et al. 2019

IEEE Trans. on Image Process.

Self Cite

282

116

View full text Add to dashboard Cite

Scene text detection is an important step of scene text reading system. The main challenges lie on significantly varied sizes and aspect ratios, arbitrary orientations and shapes. Driven by recent progress in deep learning, impressive performances have been achieved for multi-oriented text detection. Yet, the performance drops dramatically in detecting curved texts due to the limited text representation (e.g., horizontal bounding boxes, rotated rectangles, or quadrilaterals). It is of great interest to detect curved texts, which are actually very common in natural scenes. In this paper, we present a novel text detector named TextField for detecting irregular scene texts. Specifically, we learn a direction field pointing away from the nearest text boundary to each text point. This direction field is represented by an image of two-dimensional vectors and learned via a fully convolutional neural network. It encodes both binary text mask and direction information used to separate adjacent text instances, which is challenging for classical segmentationbased approaches. Based on the learned direction field, we apply a simple yet effective morphological-based post-processing to achieve the final detection. Experimental results show that the proposed TextField outperforms the state-of-the-art methods by a large margin (28% and 8%) on two curved text datasets: Total-Text and SCUT-CTW1500, respectively, and also achieves very competitive performance on multi-oriented datasets: IC-DAR 2015 and MSRA-TD500. Furthermore, TextField is robust in generalizing to unseen datasets. The code is available at https://github.com/YukangWang/TextField.

show abstract

“…There are also some other related studies. DeepFlux [ 29 ] is a method of extracting natural target skeleton. It constructs a vector field in the region of interest.…”

Section: Introductionmentioning

confidence: 99%

Road Extraction from High Resolution Remote Sensing Images Based on Vector Field Learning

Liang

Shi

Ding

et al. 2021

Sensors

View full text Add to dashboard Cite

Accurate and up-to-date road network information is very important for the Geographic Information System (GIS) database, traffic management and planning, automatic vehicle navigation, emergency response and urban pollution sources investigation. In this paper, we use vector field learning to extract roads from high resolution remote sensing imaging. This method is usually used for skeleton extraction in nature image, but seldom used in road extraction. In order to improve the accuracy of road extraction, three vector fields are constructed and combined respectively with the normal road mask learning by a two-task network. The results show that all the vector fields are able to significantly improve the accuracy of road extraction, no matter the field is constructed in the road area or completely outside the road. The highest F1 score is 0.7618, increased by 0.053 compared with using only mask learning.

show abstract

DeepFlux for Skeletons in the Wild

Cited by 44 publications

References 49 publications

Genetic Feature Fusion for Object Skeleton Detection

Genetic Feature Fusion for Object Skeleton Detection

TextField: Learning a Deep Direction Field for Irregular Scene Text Detection

Road Extraction from High Resolution Remote Sensing Images Based on Vector Field Learning

Contact Info

Product

Resources

About