AI - Patrickz (Peerasan Buranasanti)

Real-time object detection on Vehicle with YOLOv3

Testing YOLOv3: Real-Time Object Detection

just my note I testing real-time object detection around 2018. So don't compare the result with today technologies. To do this in that time need high performance GPU hardware Yolov3 - All Pre-Trained Model ComparePre-Trained Model: YOLOv3-TinyYOLOv3-SPP, Spatial Pyramid PoolingYOLOv3yolo9000Keyword: #YOLO #yolov3 #darknet #ComputerVision #Yolo9000

VideoProcessing

video stabilization

Video stabilization using VidStab and FFMPEG Vidstab is a video stabilization library which can be plugged-in with Ffmpeg and Transcode. "A video acquired using a hand-held camera or a camera mounted on a vehicle, typically suffers from undesirable shakes and jitters. Activities such as surfing, skiing, riding and walking while

AI

Facial Landmark

Facial landmark คือหาจุดสำคัญบนใบหน้า เช่น ตา คิ้ว จมูก ปาก รูปหน้า เพื่อนำไปใช้งานต่อไป เช่น นำไปตรวจสอบการกระพริบตา การเปลี่ยนหน้า ตำแหน่งหน้า วิเคราะห์รูปหน้า (กลม เรียว รี) ยกตัวอย่างพวกโปรแกรมบนมือถือ ที่เอาภาพจากกล้อง ไปใส่

AI

Real-time cartoon animation from Tiktok

"Pose Animator takes a 2D vector illustration and animates its containing curves in real-time based on the recognition result from PoseNet and FaceMesh. It borrows the idea of skeleton-based animation from computer graphics and applies it to vector characters." Git: https://lnkd.in/gezfSnU #ComputerVision #Tensorflow #ImageProcessing #VideoProcessing #machinelearning #deeplearning

AI

realtime deepfake

I try to generate new video from a photo #ComputerVision #AI #Deepfakes #videoprocessing #Avatarify

YOLO

YOLOv4 Released

YOLOv4 just released. It's improve speed and better object detection accurately Paper: YOLOv4: Optimal Speed and Accuracy of Object Detection (PDF) Source code: https://github.com/AlexeyAB/darknet Samples

video

Real-time deepfakes

What is deepfakes?Deepfakes (a portmanteau of "deep learning" and "fake") are synthetic media in which a person in an existing image or video is replaced with someone else's likeness. While the act of faking content is a not new, deepfakes leverage powerful techniques from machine learning and artificial intelligence

video

Stylizing Video

It is a new way to generate video content by get style from one or more keyframe then automatically propagates the stylization to the rest of the sequence. https://youtu.be/BpmYpj_6qhYI was inspired by the "Stylizing Video by Example" project at SIGGRAPH 2019. The paper was publish in

AI

Chatbot in my view

ในช่วงแรก Bot มีการกำหนดทางเดิน (Funnels design) ยกตัวอย่างเช่นระบบ #IVR (Interactive voice response) สำหรับคนที่นึกไม่ออก ก็เช่นเวลาโทรเข้าสำนักงาน บริการต่างๆ จะต้องทำ interact ด้วยการ กด 1 ติดต่อแผนกนั้น กด 2 เพื่อใส่รหัส แล้วตามด้วยเครื่องหมายสี

chatbot

I used to do a chatbot in 2016

Just for my note. I used to do a chatbot to use with a facebook page (AirbondStore) around 2016. Not so much, just try a little bit. Chat bot platform lists Flow XOChatfuelManychatBotsifyMeya - free 500/moapi.aiwit.aimicrosoft bot frameworkonsequelsmooch.io - free 2500/momsg.aifacebook Messenger Platformbeepboophqbotkitinit.aiI

AI

MIT Deep Learning Basic by Lex Fridman

MIT Deep Learning Basics: Introduction and Overview (14 Chapter)MIT Self-Driving Cars: State of the Art (2019) (10 Chapter)