AdapNet: Adaptive semantic segmentation in adverse environmental conditions

By Abhinav Valada, Johan Vertens, Ankit Dhall, Wolfram Burgard. 2017

213 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

BADGR: An Autonomous Self-Supervised Learning-Based Navigation System (2020, critical): “However,thesemethodstypicallydependonexistingSLAM approaches[5],[11]–[13]orhumans[14],[15]inorder toprovidethesemanticlabels,whichconsequentlymeans theseapproacheseitherinheritthelimitationsofgeometric approachesorarenotautonomouslyself-improving.”
Unknown Object Segmentation from Stereo Images (2021, background): “But are random augmentations on synthetic depth the best way for the network to decide when to rely on depth and when on RGB information at test time? Additionally, the fusion of RGB and corrupt depth data early in the network is nontrivial [5].”
EfficientPS: Efficient Panoptic Segmentation (2020, background): “Subsequently, Valada et al (2017) propose multi-scale residual units with parallel atrous convolutions with different dilation rates to efficiently learn multiscale features throughout the network without increasing the number of parameters.”
Cooperative Semantic Segmentation and Image Restoration in Adverse Environmental Conditions (2019, background): “[40] select RGB, depth and EVI (Enhanced Vegetation Index) as the modalities to build expert networks for segmenting images in adverse environments.”
Where Should I Walk? Predicting Terrain Properties From Images Via Self-Supervised Learning (2019, methodology): “Most camera-based approaches either classify different terrain types and then attach a value to every class [4], [5], [9], [12] or directly predict a manually defined”
WorkingHands: A Hand-Tool Assembly Dataset for Image Segmentation and Activity Mining (2019, methodology): “We experimented with the following semantic segmentation algorithms from the latest literature: SegNet [3], Mobile UNet [14], Full-Resolution Residual Networks (FRRN) [26], AdapNet [33], DeepLab [7, 8].”
Deep Multi-Modal Object Detection and Semantic Segmentation for Autonomous Driving: Datasets, Methods, and Challenges (2019, background): “Besides, some works extract additional sensing information, such as optical flow [119], depth [60], [124], [125], or other multi-spectral images [90], [124].”
Guided Curriculum Model Adaptation and Uncertainty-Aware Evaluation for Semantic Nighttime Image Segmentation (2019, methodology): “Contrary to these domain-specific methods, previous work also includes both methods designed for robustness to illumination changes, by employing domaininvariant representations [1, 25] or fusing information from complementary modalities and spectra [33], and datasets with adverse illumination [21, ”
Survey on semantic segmentation using deep learning techniques (2019, methodology): “Adapting the idea of ResNet-50 [34], an architecture called Adaptive network or AdapNet is proposed by Valada et al. [40].”
SMSnet: Semantic motion segmentation using deep convolutional neural networks (2017, unknown): “Therefore we utilize our previously proposed multiscale ResNet skip layers [23] in the architecture to incorporate scale invariance.”
A survey of deep learning techniques for autonomous driving (2019, methodology): “Semantic segmentation networks, like SegNet (Badrinarayanan et al., 2017), ICNet (H. Zhao et al., 2018), ENet (Paszke, Chaurasia, Kim, & Culurciello, 2016), AdapNet (Valada, Vertens, Dhall, & Burgard, 2017), or Mask RCNN (He et al., 2017), are mainly encoder–decoder architectures with a pixelwise cl”