Box2Pix: Single-Shot Instance Segmentation by Assigning Pixels to Object Boxes

Uhrig, Jonas; Rehder, Eike; Fröhlich, Björn; Franke, Uwe; Brox, Thomas

doi:10.1109/ivs.2018.8500621

Cited by 71 publications

(49 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…the 3D car center, which provides stronger supervision and helps learn more robust networks. Previously, inducing relative position of object instances has also been shown to be effective in instance segmentation [58,33]. Formally, let…”

Section: A Direct Approachmentioning

confidence: 99%

ApolloCar3D: A Large 3D Car Instance Understanding Benchmark for Autonomous Driving

Song

Wang²,

Zhou

et al. 2019

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

153

105

View full text Add to dashboard Cite

show abstract

Section: A Direct Approachmentioning

confidence: 99%

ApolloCar3D: A Large 3D Car Instance Understanding Benchmark for Autonomous Driving

Song

Wang²,

Zhou

et al. 2019

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

153

105

View full text Add to dashboard Cite

show abstract

“…Object masks are then obtained in postprocessing by assigning pixels to cluster centers in this space. The Box2Pix method [23] instead uses supervised learning to map pixels to a specific target space, namely the 2D offsets from the given pixel towards its instance's center, which are found through a bounding box detection branch.…”

Section: Related Workmentioning

confidence: 99%

“…Instance Stixels expand the idea of Semantic Stixels by additionally training a CNN to output a 2D estimation of the position of the instance center for each pixel. This estimation is predicted in image coordinates, as proposed in the Box2Pix method [23]. More specifically, the CNN predicts 2D offsets Ω p ∈ R 2 (i.e.…”

Section: B Instance Stixelsmentioning

confidence: 99%

“…As shown in [23], semantic segmentation networks can be trained to simultaneously predict class-wise probabilities and instance offsets (described in section III-C). To focus our experiments on the influence of the offset prediction, we use a separate Dilated Residual Network [21] with the same architecture (drn d 22), which was already trained on Cityscapes for semantic segmentation.…”

Section: B Training the Offset Cnnmentioning

confidence: 99%

See 1 more Smart Citation

Instance Stixels: Segmenting and Grouping Stixels into Objects

Hehn

Kooij

Gavrila

2019

2019 IEEE Intelligent Vehicles Symposium (IV)

View full text Add to dashboard Cite

State-of-the-art stixel methods fuse dense stereo and semantic class information, e.g. from a Convolutional Neural Network (CNN), into a compact representation of driveable space, obstacles, and background. However, they do not explicitly differentiate instances within the same class. We investigate several ways to augment single-frame stixels with instance information, which can similarly be extracted by a CNN from the color input. As a result, our novel Instance Stixels method efficiently computes stixels that do account for boundaries of individual objects, and represents individual instances as grouped stixels that express connectivity. Experiments on Cityscapes demonstrate that including instance information into the stixel computation itself, rather than as a post-processing step, increases Instance AP performance with approximately the same number of stixels. Qualitative results confirm that segmentation improves, especially for overlapping objects of the same class. Additional tests with ground truth instead of CNN output show that the approach has potential for even larger gains. Our Instance Stixels software is made freely available for non-commercial research purposes.

show abstract

“…Inspired by recent works [16,22,33] in general semantic instance segmentation, we aim to design a segmentation-based Single-shot Arbitrarily-Shaped Text detector (SAST), which integrates both the high-level object knowledge and low-level pixel information in a single shot and detects scene text of arbitrary shapes with high accuracy and efficiency. Employing a FCN [27] model, various geometric properties of text regions, including text center line (TCL), text border offset (TBO), text center offset (TCO), and text vertex offset (TVO), are designed to learn simultaneously under a multi-task learning formulation.…”

Section: Introductionmentioning

confidence: 99%

A Single-Shot Arbitrarily-Shaped Text Detector based on Context Attended Multi-Task Learning

Wang

Zhang

et al. 2019

Proceedings of the 27th ACM International Conference on Multimedia

View full text Add to dashboard Cite

Detecting scene text of arbitrary shapes has been a challenging task over the past years. In this paper, we propose a novel segmentationbased text detector, namely SAST, which employs a context attended multi-task learning framework based on a Fully Convolutional Network (FCN) to learn various geometric properties for the reconstruction of polygonal representation of text regions. Taking sequential characteristics of text into consideration, a Context Attention Block is introduced to capture long-range dependencies of pixel information to obtain a more reliable segmentation. In post-processing, a Point-to-Quad assignment method is proposed to cluster pixels into text instances by integrating both high-level object knowledge and low-level pixel information in a single shot. Moreover, the polygonal representation of arbitrarily-shaped text can be extracted with the proposed geometric properties much more effectively. Experiments on several benchmarks, including ICDAR2015, ICDAR2017-MLT, SCUT-CTW1500, and Total-Text, demonstrate that SAST achieves better or comparable performance in terms of accuracy. Furthermore, the proposed algorithm runs at 27.63 FPS on SCUT-CTW1500 with a Hmean of 81.0% on a single NVIDIA Titan Xp graphics card, surpassing most of the existing segmentation-based methods.

show abstract

Box2Pix: Single-Shot Instance Segmentation by Assigning Pixels to Object Boxes

Cited by 71 publications

References 22 publications

ApolloCar3D: A Large 3D Car Instance Understanding Benchmark for Autonomous Driving

ApolloCar3D: A Large 3D Car Instance Understanding Benchmark for Autonomous Driving

Instance Stixels: Segmenting and Grouping Stixels into Objects

A Single-Shot Arbitrarily-Shaped Text Detector based on Context Attended Multi-Task Learning

Contact Info

Product

Resources

About