2018_CVPR_Papers

Good deep-learning papers in 2018 IEEE Conference on Computer Vision and Pattern Recognition.

三维检测

✅ [Frustum PointNets for 3D Object Detection from RGB-D Data]

超分辨率

✅ [Enhancing the Spatial Resolution of Stereo Images using a Parallax Prior]

多类别/多任务网络

✅ [DiverseNet: When One Right Answer Is Not Enough]

✅ [Very Large-Scale Global SfM by Distributed Motion Averaging]

目标追踪

✅ [Context-aware Deep Feature Compression for High-speed Visual Tracking]

✅ [End-to-end Flow Correlation Tracking with Spatial-temporal Attention]

视频插值

✅ [Context-aware Synthesis for Video Frame Interpolation]

训练提升技巧

✅ [Between-class Learning for Image Classification]

✅ [Decorrelated Batch Normalization]

✅ [Tips and Tricks for Visual Question Answering:Learnings from the 2017 Challenge]

图像分割

✅ [Learning to Segment Every Thing]

✅ [Low-Latency Video Semantic Segmentation]

风格迁移

✅ [Arbitrary Style Transfer with Deep Feature Reshuffle]

图像除雨

✅ [Robust Video Content Alignment and Compensation for Rain Removal in a CNN Framework]

目标检测识别

✅ [R-FCN-3000 at 30fps: Decoupling Detection and Classification]

视频摘要

✅ [Reconstruction Network for Video Captioning]

理论研究

✅ [The Perception-Distortion Tradeoff]

行人重识别

✅ [Person Transfer GAN to Bridge Domain Gap for Person Re-Identification]

网络结构

✅ [CSRNet: Dilated Convolutional Neural Networks for Understanding the Highly Congested Scenes]