In artificial intelligence learning, everyone will more or less pay attention to some official accounts , but as the amount of information increases every day , the fragmented content makes everyone overwhelmed. How to mine valuable content and how to quickly access the content you need is a headache.
Therefore, Dabai will gather together the essence of the artificial intelligence field, dozens of official accounts released every week . At the same time, the classification of the system is carried out, so that everyone can understand the weekly technical trends of the artificial intelligence industry at a glance. I hope it will be helpful to everyone. If you want to view the previous weekly articles, click the link to view it. At the same time, Dabai will also extract more systematic knowledge points for each project in all weekly reports, sort them into "Summary of Dabai AI Weekly Essence" , click to view . In addition, Dabai is also constantly collecting and updating the video sharing of the authors and working experience of each project algorithm, hoping to improve the efficiency of some explorations , click to view . Finishing summary: Jiang Dabai Content cycle: 2020.09.07-2020.09.14 Synchronous public number: Jiang Dabai
table of Contents
1 Organize the list of public accounts involved
(1) I love computer vision
(2) Cver
(3) Datawhale
(4) AI algorithm and image processing
(5
) Opencv Academy (6) PaperWeekly
(7) Xinzhiyuan (8) Jishi
Platform
(9) Machine Heart
(10) Machine Learning Algorithm Engineer
(11) AI Research Institute
(12) GiantPandaCV
(13) AI Deep Learning Sight
(14) July Online Lab
(15) Artificial Intelligence Frontier Lecture
(16) AI Technology Review
(17) Machine Learning Algorithm and Python refinement
(18) Microsoft researcher AI headline
(19) VALSE
(20) AI algorithm training camp
(21) There are three AI
(22) AlWalker
(23) Qubit
(24) AI park
(25) AI artificial intelligence beginner
2 Summary of industry highlights
2.1 Basic knowledge section
2.1.1 Deep learning related
(1) Let's talk about the problem of category imbalance: the comprehensive analysis of weight adjustment and magic change Loss
Link: https://mp.weixin.qq.com/s/mKIguadoG9GtKMCPvkSZaw
2.1.2 Machine learning related
(1) One article explains feature engineering! Classic foreign language PPT and Chinese analysis
Link: https://mp.weixin.qq.com/s/GWMZ1jwbchE8O0r6EduYtQ
2.2 Research direction
2.2.1 Target detection
(1) CVPR2020 Night Target Detection Challenge Champion Solution Interpretation
Link: https://mp.weixin.qq.com/s/FVur6f05dMhmZgGb1aD-ww
(2) 1.3MB ultra-light YOLO algorithm! Universal on the whole platform, the accuracy rate is close to YOLOv3, and the speed is 45% faster丨Open source
link: https://mp.weixin.qq.com/s/-17BGx4cGme5cfwY8HRwWQ
(3) YOLO-v4 target detection, face change, video coloring all Real-time mobile terminal implementation, Northeastern University in the United States proposed a fully automatic real-time mobile terminal AI framework
Link: https://mp.weixin.qq.com/s/OEEPYvmtKONg94BWhjfdsw
(4) YOLO in the Dark See the dark night | Target detection
link in the dark night : Https://mp.weixin.qq.com/s/Geaag2pVWaasNE4NWurJaA
(5) GitHub: The most complete collection of small target detection data
Link: https://mp.weixin.qq.com/s/qiGvAW9GH7cEb7LXkuPi8Q
(6) YOLO- Fastest: Currently the fastest YOLO network! Up to 148 FPS
links on the mobile terminal : https://mp.weixin.qq.com/s/S5TleBBjGqQsjnNnlGzybA
(7) [Super fast and light YOLO] YOLO-Fastest complete graphic tutorials from Darknet source code compilation, testing and training !
Link: https://mp.weixin.qq.com/s/kHnl6jExYdiDY_M4130_sg
(8) Summarize target detection matching strategies and improvement ideas
Link: https://mp.weixin.qq.com/s/-TQLzBHDkjTFOf58-Aje6w
(9 ) Focal Loss and the man behind it RetinaNet
link: https://mp.weixin.qq.com/s/yszNSN35htZWJzOfU5Si8w
(10) [Detection + Retrieval] A model that allows you not only to see but also to find, set detection and Search and Yishen’s works
Link: https://mp.weixin.qq.com/s/4-UJ0vjbiVbUZWtlmb9tdg
2.2.2 Image classification
(1) Common tricks
links in image classification : https://mp.weixin.qq.com/s/pGOQDjd2hghF0syCl2y4tg
2.2.3 Target tracking
(1) ECCV 2020 GigaVision Challenge "Pedestrian and Vehicle Detection" and "Multi-Target Tracking" champion program interpretation
link: https://mp.weixin.qq.com/s/0xnkrlNUfW1Q1hPYPCW7OQ
2.2.4 Image quality evaluation
(1) Talk about image quality quantitative evaluation standards
Link: https://mp.weixin.qq.com/s/PkweYIAEexQb-iNY39JRXQ
2.2.5 Optical flow estimation
(1) Optical flow estimation-from traditional methods to deep learning
Link: https://mp.weixin.qq.com/s/HI_6R2HyQlB19Qunpmleiw
2.3 Project Application Section
2.3.1 Defect detection
(1) Easily learn Pytorch-migration learning to realize surface defect inspection
Link: https://mp.weixin.qq.com/s/_3oo7O6IJnoxE9ybllXpFg
(2) Interpretation of the competition plan: 8 groups of Tianchi aluminum profile surface defects finals plan, including data Collections, model improvements, innovation points, etc.
Link: https://mp.weixin.qq.com/s/D2UA7_2Tzoc4B2xZAIrnvA
2.4 Opencv board
(1) Summary | Atypical deep learning models in OpenCV4
Link: https://mp.weixin.qq.com/s/yS59HddG6UXGrtPjnG9gwQ
3 Summary of scientific research & industry dynamics
3.1 Research related
(1) Explain the improvement method of neural network in a simple way!
Link: https://mp.weixin.qq.com/s/VGbNk00AWLozyYjEUxBrNQ
(2) Let the algorithm land, some thoughts on data products
Link: https://mp.weixin.qq.com/s/y-WDLboUKMNtGAPL1dzDGg
(3 ) Summary of data analysis specifications!
Link: https://mp.weixin.qq.com/s/wJmCGs3uF95XRRiWY1UCRg
(4) How to plan and practice a PhD career: my personal experience
Link: https://mp.weixin.qq.com/s/I3RTWGVo3tsBG1by_9ToeQ
( 5) Why is deep learning not robust? An overview of the robustness of the 4D long article against resistance
Link: https://mp.weixin.qq.com/s/EyWxOnj6Aie41jMavZXWnw
(6) Detailed notes on the implementation process of the deep learning project, AI architects are full of praise
Link: https:// mp.weixin.qq.com/s/ERz6jiIAO3eUoV3iRO5Y1w
3.2 Industry News
(1) From 2012 to now, count the classic achievements in the field of deep learning over the years!
Link: https://mp.weixin.qq.com/s/kpS20oi3Fm0orwo2YMVgAQ
(2) RTX3090 era latest GPU purchase guide: Which graphics card is worthy of my alchemy furnace?
Link: https://mp.weixin.qq.com/s/v67u0V0sXN9NgwK5G-aorw
(3) Sharing of excellent open source projects this week: deep learning framework for multi-animal pose tracking, full version of facial data set and other 8 major projects
Link: https://mp.weixin.qq.com/s/wlUC0hsUIiMamZN6mutGZA
(4) She is the protagonist of the "Graduation Photo of One Person" at Peking University. She is the only seedling major in paleontology in 2010. Ten years later, she started AI
link: https:/ /mp.weixin.qq.com/s/cw0fe5w6qOT872e9aNBNYw
(5) Baidu opened the black technology conference to CCTV, and the host repeatedly sighed: liberal arts students understand
Link: https://mp.weixin.qq.com/ s/xRJqnyjkRMMtPCv76uulCw
4 pytorch basic tutorial
(1) Summary million words, the core developers a comprehensive interpretation PyTorch internal mechanism
links: https://mp.weixin.qq.com/s/DH5FX2Ue0j00dlUAfMgiJQ
(2) students should master the five kinds Pytorch parallel training methods (more with a single card)
link : Https://mp.weixin.qq.com/s/wKjDHOvu6yImZk_C8JKqVA
(3) 6 tips for using Pytorch efficiently: Provide powerful motivation for your training pipeline
Link: https://mp.weixin.qq.com/s /a1u6iCRYNwfCRWJIdv9bqQ
5 Basic knowledge
(1) Code + Example: Full solution of the "axis" in deep learning
Link: https://mp.weixin.qq.com/s/BIH3vf3y2v4rW6DK4yptIQ
