In this video, we look at instance segmentation and introduce the concepts of Mask-RCNN.
Watch on:FAU TVFAU TV (no memes)YouTube
Read the Transcript (Summer 2020) at:LMETowards Data Science
In this video, we look at some ideas on how to perform object detection really quickly. This leads to single shot detectors of which YOLO is one of the most popular ones. If you are in need of multi-scale object detection, Retina-Net is a popular choice.
Watch on:FAU TVFAU TV (no memes)YouTube
...
In this video, we start looking into object detection. We start with classical ideas, re-visit the concept of a fully convolutional neural network, and start developing a fast regional CNN detector which finally leads to Faster RCNN.
Watch on:FAU TVFAU TV (no memes)YouTube
Read the Transcript...
In this video, we discuss ideas on how to improve on image segmentation. In particular skip connections as used in the U-Net have been applied very successfully here. Also, we, look into other advanced methods such as stacked hourglasses, convolutional pose machines, and conditional random fields i...
In this video, we introduce the concepts of segmentation and object detection. For image segmentation, you use a CNN encoder in combination with a CNN decoder. We introduce several concepts on how to perform the upsampling in der decoder.
Watch on:FAU TVFAU TV (no memes)YouTube
Read the Trans...
In this last video on unsupervised learning, we introduce some more advanced GAN concepts to avoid mode collapse and strong intra-batch correlation using virtual batch normalization, unrolled GANs, and minibatch discrimination.
Watch on:FAU TVFAU TV (no memes)YouTube
Read the Transcript (Summ...
In this video, we talk about conditional GANs and the CycleGAN.
Watch on:FAU TVFAU TV (no memes)YouTube
Read the Transcript (Summer 2020) at:LMETowards Data Science
In this video, we talk about the basic ideas of Generative Adversarial Networks (GANs) and show some examples.
Watch on:FAU TVFAU TV (no memes)YouTube
Read the Transcript (Summer 2020) at:LMETowards Data Science
In this video, we show fundamental concepts of autoencoders (AEs) ranging from undercomplete and sparse AEs, over stacked and denoising AEs all the way to Variational Autoencoders.
Watch on:FAU TVFAU TV (no memes)YouTube
Read the Transcript (Summer 2020) at:LMETowards Data Science
In this video, we introduce the topic of unsupervised learning and start looking at Restricted Boltzmann Machines as one of the first approaches for Unsupervised Deep Learning.
Watch on:FAU TVFAU TV (no memes)YouTube
Read the Transcript (Summer 2020) at:LMETowards Data Science