Segmenting Transparent Objects in the Wild

Xie, Enze; Wang, Wenjia; Wang, Wenhai; Ding, Mingyu; Shen, Chunhua; Luo, Ping

doi:10.1007/978-3-030-58601-0_41

Cited by 104 publications

(110 citation statements)

References 22 publications

Supporting

Mentioning

110

Contrasting

Order By: Relevance

“…Datasets for the segmentation of transparent objects in real-world images have been mostly created by manual annotation and image mating. 43–46,57 The largest of these datasets is Trans10k, 45 with 10k images in which the region of the transparent object is marked. The LabPics dataset contains 8k images of mostly transparent vessels in labs, hospitals, and other settings.…”

Section: Related Workmentioning

confidence: 99%

Predicting 3D shapes, masks, and properties of materials inside transparent containers, using the TransProteus CGI dataset

Eppel

Wang

et al. 2022

Digital Discovery

View full text Add to dashboard Cite

show abstract

Section: Related Workmentioning

confidence: 99%

Predicting 3D shapes, masks, and properties of materials inside transparent containers, using the TransProteus CGI dataset

Eppel

Wang

et al. 2022

Digital Discovery

View full text Add to dashboard Cite

show abstract

“…Recently, Vision Transformer (ViT) has been used more and more in the field of computer vision, such as object detection [ 35 , 42 ], image classification [ 43 ] and semantic segmentation [ 44 ]. Inspired by SuperGlue, Sun et al proposed the LoFTR [ 45 ] algorithm.…”

Section: Related Workmentioning

confidence: 99%

LFM: A Lightweight LCD Algorithm Based on Feature Matching between Similar Key Frames

Zhu

Liu

et al. 2021

Sensors

View full text Add to dashboard Cite

Loop Closure Detection (LCD) is an important technique to improve the accuracy of Simultaneous Localization and Mapping (SLAM). In this paper, we propose an LCD algorithm based on binary classification for feature matching between similar images with deep learning, which greatly improves the accuracy of LCD algorithm. Meanwhile, a novel lightweight convolutional neural network (CNN) is proposed and applied to the target detection task of key frames. On this basis, the key frames are binary classified according to their labels. Finally, similar frames are input into the improved lightweight feature matching network based on Transformer to judge whether the current position is loop closure. The experimental results show that, compared with the traditional method, LFM-LCD has higher accuracy and recall rate in the LCD task of indoor SLAM while ensuring the number of parameters and calculation amount. The research in this paper provides a new direction for LCD of robotic SLAM, which will be further improved with the development of deep learning.

show abstract

“…This method overcomes the feature misalignment problem of previous methods based on regression to a large extent and improves the performance significantly. In addition, [50] proposes a new method of using Transformer to segment transparent objects in the field, which not only proposes a new fine-grained transparent object segmentation dataset, but also proposes a new segmented pipeline Trans2Seg based on Transformer. First, Trans2Seg's Transformer encoder provides a global acceptance domain instead of CNN's local acceptance domain.…”

Section: Future Prospectsmentioning

confidence: 99%

Two-Stream Convolutional Neural Network for Video Action Recognition

2021

KSII TIIS

View full text Add to dashboard Cite

Segmenting Transparent Objects in the Wild

Cited by 104 publications

References 22 publications

Predicting 3D shapes, masks, and properties of materials inside transparent containers, using the TransProteus CGI dataset

Predicting 3D shapes, masks, and properties of materials inside transparent containers, using the TransProteus CGI dataset

LFM: A Lightweight LCD Algorithm Based on Feature Matching between Similar Key Frames

Two-Stream Convolutional Neural Network for Video Action Recognition

Contact Info

Product

Resources

About