Computer Vision Machine Learning Deep Learning CNNs

A Gaussian mixture model is a probabilistic model that assumes all the data points are generated from a mixture of a finite number of Gaussian distributions with unknown parameters. One can think of mixture models as generalizing k-means clustering to incorporate information about the covariance structure of the data as well as the centers of the latent Gaussians. Linear regression  is an analysis that assesses whether one or more predictor variables explains the dependent (criterion) variable.  The regression has five key assumptions: Linear relationship Multivariate normality No or little multicollinearity No auto-correlation Homoscedasticity In probability theory, a  Markov model  is a stochastic  model  used to  model  randomly changing systems where it is assumed that future states depend only on the current state not on the events that occurred before it A  hidden Markov model  is a Markov chain for which the state is only partially observable. In other words,

1 What's your biggest weakness? My biggest weakness is that when I am working on one task. It's very hard to take my thoughts out from the current task. When I try to shift myself to another task, all the thoughts in my head are on the previous task. I need to spend longer time to adjust myself to the new task. So at one time, it's very hard for me to work with multi-task. At a conversation with my friend, she told me she has the same problem, and her solution is that she work on one task for three days and then shift to another one for several days. The strategy is not to shift too frequently, which also works for me. 2 What's your most challenging time in your Ph.D.? In the beginning of my fourth year of Ph.D. program. Almost all the Ph.D. students were enrolled with me in the same year has published one or two papers, but I don't have any because my experiments were always not as good as the baseline. I am very struggled and even depressed. I begin to think s

Customer Obsession Leaders start with the customer and work backward. They work vigorously to earn and keep customer trust. Although leaders pay attention to competitors, they obsess over customers. 1 In the presentation: In the amazon store, if we can know the expression of our customer when they are buying some products, then we can have a better understanding the relationship between our customer and this product. For example, based the happiness degree of their expression, we can the customer like it very much, or just normally like it, or just need it.  2 My own example: In my research, I also have many customers: my advisor, my colleagues (labmates), and the people who read my paper. To my advisor, I should satisfy my advisor, when my idea is conflicted with my advisor, or we have the different opinion for some problems. Of course, we will communicate and each other. I will respect her idea and follow her instructions. At the same time, I will also implement my own

This paper proposed a CNN framework for object instance segmentation. The method is called Mask R-CNN, which is the extension of faster R-CNN by adding an FCN branch for predicting segmentation masks on each RoI.

SSD is the extension or updating version of YOLO. Compared with YOLO: 1 SSD use the multi-scale feature maps for detection, but the YOLO only use one scale for detection. 2 Before the decision layer, the YOLO use the fully connected layer for detection, but the SSD use the convolution filter for detection, which can improve the final detection performance. 3 Default boxes and aspect ratios. there are a set of default bounding boxes with each feature map cell for multiple feature maps at the top of the network. R-CNN: Selective Search Region Proposals; SVM classification Fast R-CNN: Selective Search Region Proposals; Shared computation by using SSP. Faster R-CNN: Region Proposal Network OverFeat VS SSD: If the SSD only use one default box per location from the topmost feature map, the SSD would have similar architecture to OverFeat. YOLO VS SSD: If use the whole topmost feature map and add a fully-connected layer for predictions instead of convolutional predictors, it'

Title: You Only Look Once: Unified, Real-Time Object Detection YOLO is a new approach for object detection. You only need run a single CNN on the image once and make the decision for each grid. Instead of using the sliding windows or region proposes. the YOLO proposed to divide the image to S*S grid and predict each grid with regression. Contribution: The traditional model will do a classification for each proposal or sliding window, but the proposed YOLO divide the image to S*S grid and train a regression model. The ground truth target vector has the dimension of S*S*(B*5+C). Each grid cell predicts B bounding boxes and confidence scores for those boxes. The confidence score reflects how confidence the current grid contain the object and also how accurate it thinks the box is that it predicts. Network: The network has 24 constitutional layers followed by 2 fully connected layers. There are four 2*2 max-pooling layers with stride 2. Instead of inception modules used by Goog

Amazon Go Project Analysis What's is Amazon Go? Amazon Go is a new kind of grocery store with no checkout required. Amazon proposed the Checkout-free Shopping experience. With the Amazon Go app to enter the store, grab the products you want, and just go! No line, no checkout. December 5, 2016, Amazon publish a video about the Amazon Go. Four years ago, we started to wonder, what would shopping look like if you could walk into a store, grab what you want, and just go? What if we could weave the most advanced machine learning, computer vision and AI into the very fabric of a store, so you never have to wait in line? No lines, no checkouts, no registers. … Use the Amazon Go app to enter, then put away your phone and start shopping. Take whatever you like. Anything you pick up is automatically added to your virtual cart. If you change your mind about that cupcake, just put it back. Our technology will update your virtual cart automatically. So how does it work? We us