The image auto rectification project at Google aims to create a pleasanter version of user photos by correcting the small, involuntary camera rotations (roll / pitch/ yaw) that often occur in non-professional photographs. Our system takes the image closer to the fronto-parallel view by performing an affine rectification on the image that restores parallelism of lines that are parallel in the fronto-parallel image view. This partially corrects perspective distortions, but falls short of full metric rectification which also restores angles between lines. On the other hand the 2D homography for our rectification can be computed from only two (as opposed to three) estimated vanishing points, allowing us to fire upon many more images. A new RANSAC based approach to vanishing point estimation has been developed. The main strength of our vanishing point detector is that it is line-less, thereby avoiding the hard, binary (line/no-line) upstream decisions that cause traditional algorithm to ignore much supporting evidence and/or admit noisy evidence for vanishing points. A robust RANSAC based technique for detecting horizon lines in an image is also proposed for analyzing correctness of the estimated rectification. We post-multiply our affine rectification homography with a 2D rotation which aligns the closer vanishing point with the image Y axis.
In this paper, we present a uni ed end-to-end approach to build a large scale Visual Search and Recommendation system for ecommerce. Previous works have targeted these problems in isolation. We believe a more e ective and elegant solution could be obtained by tackling them together. We propose a uni ed Deep Convolutional Neural Network architecture, called VisNet 1 , to learn embeddings to capture the notion of visual similarity, across several semantic granularities. We demonstrate the superiority of our approach for the task of image retrieval, by comparing against the state-of-the-art on the Exact Street2Shop [14] dataset. We then share the design decisions and trade-o s made while deploying the model to power Visual Recommendations across a catalog of 50M products, supporting 2K queries a second at Flipkart, India's largest e-commerce company. e deployment of our solution has yielded a signi cant business impact, as measured by the conversion-rate.
Accurate demand forecasts can help on-line retail organizations better plan their supply-chain processes. The challenge, however, is the large number of associative factors that result in large, non-stationary shifts in demand, which traditional time series and regression approaches fail to model. In this paper, we propose a Neural Network architecture called AR-MDN, that simultaneously models associative factors, time-series trends and the variance in the demand. We first identify several causal features and use a combination of feature embeddings, MLP and LSTM to represent them. We then model the output density as a learned mixture of Gaussian distributions. The AR-MDN can be trained end-to-end without the need for additional supervision. We experiment on a dataset of an year's worth of data over tens-of-thousands of products from Flipkart. The proposed architecture yields a significant improvement in forecasting accuracy when compared with existing alternatives.
Abstract-A new computational model for optical flow estimation isproposed. The proposed model utilizes trajectory information present in a multiframe spatio-temporal volume. Optical flow estimation is formulated as an optimization problem. The solution to this optimization problem yields a velocity field corresponding to smoothest and shortest trajectories of constant intensity points within the spatio-temporal volume. The approach is motivated by principles of inertia of morion and least action in physics and vision psychology. An analogy between a trajectory and a "thin wire" is discussed. A simple mechanism for handling trajectory discontinuities is also incorporated. The optimization problem is solved by stochastic relaxation techniques. Some experimental results and performance comparisons with two existing optical flow estimation techniques are presented to demonstrate the effectiveness of the proposed approach.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.