【深度学习】目标检测的一些学习资源

本文探讨了形变卷积网络的各种应用及改进,包括最新版本的Deformable ConvNets v2,并讨论了轻量级神经网络的设计理念和发展趋势,如MobileNet及其变体。

摘要生成于 C知道 ,由 DeepSeek-R1 满血版支持, 前往体验 >

NMS也能玩出花样来
https://zhuanlan.zhihu.com/p/28129034?utm_source=qq&utm_medium=social&utm_oi=62863464988672

nms
https://zhuanlan.zhihu.com/p/51008736?utm_source=qq&utm_medium=social&utm_oi=62863464988672

test Soft-NMS with a state-of-the-art detector, so Deformable-R-FCN was trained on 800x1200 size images with 15 anchors.
https://github.com/bharatsingh430/Deformable-ConvNets

形变卷积
https://m.weibo.cn/status/4313255994165020?wm=3333_2001&from=108B093010&sourcetype=qq&featurecode=newtitle

Deformable Convolutional Networks v2 with Pytorch
https://m.weibo.cn/status/4314335885481850?wm=3333_2001&from=108B093010&sourcetype=qq&featurecode=newtitle

Appearance and Pose-Conditioned Human Image Generation using Deformable GANs
https://m.weibo.cn/status/4367735477196987?wm=3333_2001&from=108B093010&sourcetype=qq&featurecode=newtitle

Deformable ConvNets v2: More Deformable, Better Results》X Zhu, H Hu, S Lin, J Dai [University of Science and Technology of China & Microsoft Research Asia] (2018)
https://m.weibo.cn/status/4312647262169631?wm=3333_2001&from=108B093010&sourcetype=qq&featurecode=newtitle
【Boosting算法大比拼:XGBoost vs. LightGBM vs. Catboost】
https://m.weibo.cn/status/4389595166609356?wm=3333_2001&from=108B093010&sourcetype=qq&featurecode=newtitle
用编译器、CPU pinning压榨xgboost/LightGBM最佳性能】
https://m.weibo.cn/status/4201907717510099?wm=3333_2001&from=108B093010&sourcetype=qq&featurecode=newtitle
Repo-2018 - Deep Learning Summer School + Tensorflow + OpenCV cascade training + YOLO + COCO + CycleGAN + AWS EC2 Setup + AWS IoT Project + AWS SageMaker + AWS API Gateway + Raspberry Pi3 Ubuntu Core + Brain Waves Reconstruction’
https://m.weibo.cn/status/4298391796521561?wm=3333_2001&from=108B093010&sourcetype=qq&featurecode=newtitle
A Deep Cascade of Convolutional Neural Networks for MR Image Reconstruction
https://m.weibo.cn/status/4084700223498487?wm=3333_2001&from=108B093010&sourcetype=qq&featurecode=newtitle
A Deep Cascade of Convolutional Neural Networks for MR Image Reconstruction
https://m.weibo.cn/status/4021833525775004?wm=3333_2001&from=108B093010&sourcetype=qq&featurecode=newtitle
如何处理非均衡数据
https://m.weibo.cn/status/4344677468015493?wm=3333_2001&from=108B093010&sourcetype=qq&featurecode=newtitle

七招处理非均衡数据
https://m.weibo.cn/status/4114562032238642?wm=3333_2001&from=108B093010&sourcetype=qq&featurecode=newtitle
仅需一行代码(torch.hub.load())复用ResNet, ResNext, BERT, GPT, PGAN, Tacotron, DenseNet, MobileNet等最新模型
https://m.weibo.cn/status/4383204880022972?wm=3333_2001&from=108B093010&sourcetype=qq&featurecode=newtitle

https://m.weibo.cn/status/4381889060447517?wm=3333_2001&from=108B093010&sourcetype=qq&featurecode=newtitle
YOLOv3的Keras实现(Mobilenetv1/VGG16/ResNet101/ResNeXt101
https://m.weibo.cn/status/4366490474622355?wm=3333_2001&from=108B093010&sourcetype=qq&featurecode=newtitle
MobileNet及其变体为什么这么快
https://m.weibo.cn/status/4362308119957312?wm=3333_2001&from=108B093010&sourcetype=qq&featurecode=newtitle
OpenCV/OpenPose MobileNet人体姿态估计
https://m.weibo.cn/status/4341864779857839?wm=3333_2001&from=108B093010&sourcetype=qq&featurecode=newtitle
2019科技发展大趋势:专业AI芯片竞赛、边缘计算成为主流、AI监管与审查、机器人过程自动化
https://m.weibo.cn/status/4322442698956161?wm=3333_2001&from=108B093010&sourcetype=qq&featurecode=newtitle
关键点对目标检测
https://m.weibo.cn/status/4270297202772451?wm=3333_2001&from=108B093010&sourcetype=qq&featurecode=newtitle
DeepFace - 基于开源框架实现的人脸识别、脸脸检测、人脸关键点检测等任务" by Riwei Chen GitHub
https://m.weibo.cn/status/3983734363139256?wm=3333_2001&from=108B093010&sourcetype=qq&featurecode=newtitle
PyTorch实现的C3D, R3D, R2Plus1D视频行为识别
https://m.weibo.cn/status/4337530109247446?wm=3333_2001&from=108B093010&sourcetype=qq&featurecode=newtitle
CV-arXiv-Daily - 分享计算机视觉每天的arXiv文章
https://m.weibo.cn/status/4334997156726990?wm=3333_2001&from=108B093010&sourcetype=qq&featurecode=newtitle
(NIST)视频扩展行为识别(ActEV)挑战
https://m.weibo.cn/status/4331512696181113?wm=3333_2001&from=108B093010&sourcetype=qq&featurecode=newtitle
ActivityNet视频行为识别挑战第三名方案分享
https://m.weibo.cn/status/4152279768653999?wm=3333_2001&from=108B093010&sourcetype=qq&featurecode=newtitle
Random Dilation Networks视频行为识别
https://m.weibo.cn/status/4135804727676272?wm=3333_2001&from=108B093010&sourcetype=qq&featurecode=newtitle
Python)泰坦尼克数据集集成/堆叠算法实践入门
https://m.weibo.cn/status/4077472565280563?wm=3333_2001&from=108B093010&sourcetype=qq&featurecode=newtitle
腾讯广告算法大赛初赛第一名方案
https://m.weibo.cn/status/4394769025166564?wm=3333_2001&from=108B093010&sourcetype=qq&featurecode=newtitle
轻量级人脸检测
https://m.weibo.cn/status/4394587303343489?wm=3333_2001&from=108B093010&sourcetype=qq
基于NAS的特征金字塔网络
https://m.weibo.cn/status/4392414737677410?wm=3333_2001&from=108B093010&sourcetype=qq&featurecode=newtitle
10个梯度下降优化算法+备忘单
https://m.weibo.cn/status/4394732019240440?wm=3333_2001&from=108B093010&sourcetype=qq&featurecode=newtitle
从MobileNet看轻量级神经网络的发展
https://media.weibo.cn/article?id=2309404388180525738877
迁移学习
http://nooverfit.com/wp/cvpr2019好的模型,迁移学习效果就更好吗?google-brain最新结/

为什么MobileNet及其变体(比如ShuffleNet)会怎么快呢
https://mp.weixin.qq.com/s/R427cUT7sryMRFMmy97QVA
MobileNet及其变体为什么这么快
https://m.weibo.cn/status/4362308119957312?wm=3333_2001&from=108B093010&sourcetype=qq&featurecode=newtitle

纵览轻量化卷积神经网络
https://m.weibo.cn/status/4315463187935347?wm=3333_2001&from=108B093010&sourcetype=qq&featurecode=newtitle
一个应用于物体识别的迁移学习工具
https://m.leiphone.com/news/201804/AOyKhuR3L0xsBUbN.html
BERT模型从训练到部署’ by xmxoxo GitHub
https://m.weibo.cn/status/4387838214458913?wm=3333_2001&from=108B093010&sourcetype=qq&featurecode=newtitle

Deep Compression/Acceleration:模型压缩加速论文汇总
https://mp.weixin.qq.com/s/cQfEAQMH8NJvvfXceaJ3Ng

评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值