Vision7 SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation We present a novel and practical deep fully convolutional neural network architecture for semantic pixel-wise segmentation termed SegNet. This core trainable segmentation engine consists of an encoder network, a corresponding decoder network followed by a arxiv.org Sementic Segmentation 이란 위 그림(오른쪽)과 같이 이미지 혹은 영상 속 .. 2022. 1. 19. SENet(Squeeze-and-Excitation Networks) SENet Squeeze-and-Excitation Networks The central building block of convolutional neural networks (CNNs) is the convolution operator, which enables networks to construct informative features by fusing both spatial and channel-wise information within local receptive fields at each layer. A broa arxiv.org 어떤 문제를 맞닥뜨렸을때, 그리고 그 문제를 머신러닝으로 해결하려고 하면 아직까지도 막연하게 '잘 되지 않을까?'라는 생각이 조금은 남아있는 것 같습니다. 하지만 실제.. 2022. 1. 5. 1x1 Convolutions & Inception Module 직장을 다니다보니 이전에 봤던것들도 기억이 안나서 정리를 다시 해보려고 합니다. 다양한 네트워크가 쏟아져나오면서 컨볼루션 연산도 기본적인 stride, filter size 등을 조절하는 것 뿐만 아니라 그룹을 지어 하는 Grouped Convolution 등 여러 방안들이 제안되었습니다. 이런 제안들 중 오늘 정리하려는 연산은 1x1(Point-Wise Convolution) 입니다. 1x1 Convolution을 생각하면 이걸 왜 하느냐에 대한 의문이 생길 수 있습니다. 첫 그림처럼 1x1 Convolution은 scalar 곱이나 마찬가지이기 때문입니다. Filter의 채널이 증가해도 output channel은 filter의 개수가 결정하기 때문에 1채널짜리 output이 생기게됩니다. 즉 Fil.. 2022. 1. 4. Pseudo-LiDAR from Visual Depth Estimation:Bridging the Gap in 3D Object Detection for Autonomous Driving Pseudo-LiDAR from Visual Depth Estimation: Bridging the Gap in 3D Object Detection for Autonomous Driving 3D object detection is an essential task in autonomous driving. Recent techniques excel with highly accurate detection rates, provided the 3D input data is obtained from precise but expensive LiDAR technology. Approaches based on cheaper monocular or stere arxiv.org 자율주행은 굉장히 많은 기술들이 합쳐진 하나의.. 2021. 11. 21. Introduction - Attention Sequence to Sequence Learning with Neural Networks Deep Neural Networks (DNNs) are powerful models that have achieved excellent performance on difficult learning tasks. Although DNNs work well whenever large labeled training sets are available, they cannot be used to map sequences to sequences. In this pap arxiv.org Neural Machine Translation by Jointly Learning to Align and Translate Neural mac.. 2021. 10. 31. Complex-YOLO: Real-time 3D Object Detection on Point Clouds Complex-Yolo Complex-YOLO: Real-time 3D Object Detection on Point Clouds Lidar based 3D object detection is inevitable for autonomous driving, because it directly links to environmental understanding and therefore builds the base for prediction and motion planning. The capacity of inferencing highly sparse 3D data in real-time arxiv.org Complex YOLO — 3D point clouds bounding box detection and t.. 2021. 9. 12. 이전 1 2 다음
