Dilated Residual Networks

Yu, Fisher; Koltun, Vladlen; Funkhouser, Thomas

doi:10.1109/cvpr.2017.75

Cited by 1,733 publications

(1,194 citation statements)

References 19 publications

Supporting

Mentioning

1,190

Contrasting

Unclassified

Order By: Relevance

“…Architectures and Training: We implement three architectures: an LSTM [48], a Transformer [49], and a dilated residual network (ResNet) [50]. We use a 12-layer Transformer with a hidden size of 512 units and 8 attention heads, leading to a 38M-parameter model.…”

Section: Models and Experimental Setupmentioning

confidence: 99%

Evaluating Protein Transfer Learning with TAPE

Rao

Bhattacharya

Thomas

et al. 2019

Preprint

548

932

View full text Add to dashboard Cite

Protein modeling is an increasingly popular area of machine learning research. Semi-supervised learning has emerged as an important paradigm in protein modeling due to the high cost of acquiring supervised protein labels, but the current literature is fragmented when it comes to datasets and standardized evaluation techniques. To facilitate progress in this field, we introduce the Tasks Assessing Protein Embeddings (TAPE), a set of five biologically relevant semi-supervised learning tasks spread across different domains of protein biology. We curate tasks into specific training, validation, and test splits to ensure that each task tests biologically relevant generalization that transfers to real-life scenarios. We benchmark a range of approaches to semi-supervised protein representation learning, which span recent work as well as canonical sequence learning techniques. We find that self-supervised pretraining is helpful for almost all models on all tasks, more than doubling performance in some cases. Despite this increase, in several cases features learned by self-supervised pretraining still lag behind features extracted by state-of-the-art non-neural techniques. This gap in performance suggests a huge opportunity for innovative architecture design and improved modeling paradigms that better capture the signal in biological sequences. TAPE will help the machine learning community focus effort on scientifically relevant problems. Toward this end, all data and code used to run these experiments are available at https://github.com/songlab-cal/tape. * Equal Contribution

show abstract

Section: Models and Experimental Setupmentioning

confidence: 99%

Evaluating Protein Transfer Learning with TAPE

Rao

Bhattacharya

Thomas

et al. 2019

Preprint

548

932

View full text Add to dashboard Cite

show abstract

“…The encoder E uses residual connections and dilated convolutions (dilation rate = 2) to enlarge the size of receptive field while preserving the spatial resolution for dense predictions [46]. Let {Ck, Rk, Dk} denote a convolutional layer, a residual block and a dilated residual block with k channels, respectively.…”

Section: Network Configurations and Implementation Detailsmentioning

confidence: 99%

Unsupervised Bidirectional Cross-Modality Adaptation via Deeply Synergistic Image and Feature Alignment for Medical Image Segmentation

Chen

Dou

Chen

et al. 2020

IEEE Trans. Med. Imaging

331

185

View full text Add to dashboard Cite

Unsupervised domain adaptation has increasingly gained interest in medical image computing, aiming to tackle the performance degradation of deep neural networks when being deployed to unseen data with heterogeneous characteristics. In this work, we present a novel unsupervised domain adaptation framework, named as Synergistic Image and Feature Alignment (SIFA), to effectively adapt a segmentation network to an unlabeled target domain. Our proposed SIFA conducts synergistic alignment of domains from both image and feature perspectives. In particular, we simultaneously transform the appearance of images across domains and enhance domain-invariance of the extracted features by leveraging adversarial learning in multiple aspects and with a deeply supervised mechanism. The feature encoder is shared between both adaptive perspectives to leverage their mutual benefits via end-to-end learning. We have extensively evaluated our method with cardiac substructure segmentation and abdominal multi-organ segmentation for bidirectional crossmodality adaptation between MRI and CT images. Experimental results on two different tasks demonstrate that our SIFA method is effective in improving segmentation performance on unlabeled target images, and outperforms the state-of-the-art domain adaptation approaches by a large margin.

show abstract

“…7 is inspired by the refinement procedures proposed in CRL [17], iResNet [12], StereoNet [8], and ActiveStere-oNet [31]. We adopted the basic architecture for refinement as described in StereoNet [8] with dilated residual blocks [28] to increase the receptive field of filtering without compromising resolution. This technique was also adopted in recent work on optical flow prediction Pwc-net [23].…”

Section: Dilation In Cost Filteringmentioning

confidence: 99%

StereoDRNet: Dilated Residual StereoNet

Chabra

Straub

Sweeney

et al. 2019

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

View full text Add to dashboard Cite

Figure 1: StereoDRNet enables estimation of high quality depth maps that opens the door to high quality reconstruction by passive stereo video. In this figure we compare the output from dense reconstruction [15] built form depth maps generated by StereoDRNet, PSMNet [2] and a structured light system [25] (termed Ground Truth). We report and visualize point-toplane distance RMS error on the reconstructed meshes with respect to the ground truth demonstrating the improvement in reconstruction over the state-of-the-art. AbstractWe propose a system that uses a convolution neural network (CNN) to estimate depth from a stereo pair followed by volumetric fusion of the predicted depth maps to produce a 3D reconstruction of a scene. Our proposed depth refinement architecture, predicts view-consistent disparity and occlusion maps that helps the fusion system to produce geometrically consistent reconstructions. We utilize 3D dilated convolutions in our proposed cost filtering network that yields better filtering while almost halving the computational cost in comparison to state of the art cost filtering architectures. For feature extraction we use the Vortex Pooling architecture [26]. The proposed method achieves state of the art results in KITTI 2012, KITTI 2015 and ETH 3D stereo benchmarks. Finally, we demonstrate that our system is able to produce high fidelity 3D scene reconstructions that outperforms the state of the art stereo system.

show abstract

Dilated Residual Networks

Cited by 1,733 publications

References 19 publications

Evaluating Protein Transfer Learning with TAPE

Evaluating Protein Transfer Learning with TAPE

Unsupervised Bidirectional Cross-Modality Adaptation via Deeply Synergistic Image and Feature Alignment for Medical Image Segmentation

StereoDRNet: Dilated Residual StereoNet

Contact Info

Product

Resources

About