Convolutional Scale Invariance for Semantic Segmentation

Krešo, Ivan; Čaušević, Denis; Krapac, Josip; Šegvić, Siniša

doi:10.1007/978-3-319-45886-1_6

Cited by 46 publications

(17 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In Table 2, we report the results evaluated on the Cityscapes test set and the comparisons with the other state-of-art systems. LDFNet achieves a 71.3% mIoU score without any pretrained model and surpasses all the other methods [10,16] designed for RGB-D semantic segmentation on this benchmark. Moreover, in Table 3, LDFNet outperforms several stateof-art networks for the RGB semantic segmentation task in terms of efficiency, such as DeepLab [2] and PSPNet [20].…”

Section: Evaluation Resultsmentioning

confidence: 85%

Incorporating Luminance, Depth and Color Information by a Fusion-Based Network for Semantic Segmentation

Hung

Hang

2019

2019 IEEE International Conference on Image Processing (ICIP)

View full text Add to dashboard Cite

Semantic segmentation has made encouraging progress due to the success of deep convolutional networks in recent years. Meanwhile, depth sensors become prevalent nowadays; thus, depth maps can be acquired more easily. However, there are few studies that focus on the RGB-D semantic segmentation task. Exploiting the depth information effectiveness to improve performance is a challenge. In this paper, we propose a novel solution named LDFNet, which incorporates Luminance, Depth and Color information by a fusion-based network. It includes a sub-network to process depth maps and employs luminance images to assist the depth information in processes. LDFNet outperforms the other state-of-art systems on the Cityscapes dataset, and its inference speed is faster than most of the existing networks. The experimental results show the effectiveness of the proposed multi-modal fusion network and its potential for practical applications.

show abstract

Section: Evaluation Resultsmentioning

confidence: 85%

Incorporating Luminance, Depth and Color Information by a Fusion-Based Network for Semantic Segmentation

Hung

Hang

2019

2019 IEEE International Conference on Image Processing (ICIP)

View full text Add to dashboard Cite

show abstract

“…The resolution of images and segmentation labels is 1242 × 375 pixels, and the type and number of segmentation class are identical to those of CamVid. For KITTI, however, the segmentation label of the test set is not disclosed publicly; thus, the KITTI database consisting of 445 images and segmentation labels provided in a previous study [32] was in our study. This database was also divided into two subsets, and the two-fold cross validation was conducted to measure the average value.…”

Section: ) Daytime Camvid and Kitti Databasesmentioning

confidence: 99%

Modified Perceptual Cycle Generative Adversarial Network-Based Image Enhancement for Improving Accuracy of Low Light Image Segmentation

et al. 2021

View full text Add to dashboard Cite

In recent years, the importance of the semantic segmentation field has been increasingly emphasized because autonomous vehicle and artificial intelligence (AI)-based robot technology are being researched extensively; and methods for accurately recognizing objects are required. Previous state-of-theart segmentation methods have been proven to be effective for databases obtained during daytime. However, in extremely low light or nighttime environments, the shape and color information of objects are very small or disappear due to an insufficient amount of external light, which makes it difficult to train the segmentation network and significantly degrades performance. In our previous work, segmentation performance in a low light environment was improved using the enhancement-based segmentation method. However, low light images could not be restored precisely and segmentation performance improvement was limited because only per-pixel loss functions were used when training the enhancement network. To overcome these drawbacks, we propose a low light image segmentation method based on a modified perceptual cycle generative adversarial network (CycleGAN). Perceptual image enhancement was performed using our network, which significantly improved segmentation performance. Unlike the existing perceptual loss, the Euclidean distance of the feature maps extracted from the pretrained segmentation network was used.In our experiments, we used low light databases generated from two famous road scene open databases, which are Cambridge-driving Labeled Video Database (CamVid) and Karlsruhe Institute of Technology and Toyota Technological Institute at Chicago (KITTI), and confirmed that our proposed method shows better segmentation performance in extremely low light environments than the existing state-of-the art methods. INDEX TERMSSemantic segmentation; Low light; Modified perceptual CycleGAN; Perceptual loss; Road scene open database.

show abstract

“…However, this database does not provide a ground-truth label for the test set. In our study, a total of 445 RGB images and ground-truth labels from the KITTI database provided in a previous study [33] were used. In the same way as the above database, this database was divided into two subsets of 223 and 222 images for a two-fold crossvalidation experiment.…”

Section: A Experimental Databases 1) Camvid and Kitti Databasesmentioning

confidence: 99%

Semantic Segmentation With Low Light Images by Modified CycleGAN-Based Image Enhancement

et al. 2020

View full text Add to dashboard Cite

In recent years, the importance of semantic segmentation has been widely recognized and the field has been actively studied. The existing state-of-the-art segmentation methods show high performance for bright and clear images. However, in low light or nighttime environments, images are blurred and noise increases due to the nature of the camera sensor, which makes it very difficult to perform segmentation for various objects. For this reason, there are few previous studies on multi-class segmentation in low light or nighttime environments. To address this challenge, we propose a modified cycle generative adversarial network (CycleGAN)-based multi-class segmentation method that improves multi-class segmentation performance for low light images. In this study, we used low light databases generated by two road scene open databases that provide segmentation labels, which are the Cambridge-driving labeled video database (CamVid) and Karlsruhe Institute of Technology and Toyota Technological Institute at Chicago (KITTI) database. Consequently, the proposed method showed superior segmentation performance compared with the other state-of-the-art methods.

show abstract

Convolutional Scale Invariance for Semantic Segmentation

Cited by 46 publications

References 20 publications

Incorporating Luminance, Depth and Color Information by a Fusion-Based Network for Semantic Segmentation

Incorporating Luminance, Depth and Color Information by a Fusion-Based Network for Semantic Segmentation

Modified Perceptual Cycle Generative Adversarial Network-Based Image Enhancement for Improving Accuracy of Low Light Image Segmentation

Semantic Segmentation With Low Light Images by Modified CycleGAN-Based Image Enhancement

Contact Info

Product

Resources

About