arXiv subjects

Kyle Genova

Publications and source records attributed to Kyle Genova.

24 records · Page 2Linked to original sources

arXiv2020

Towards Fairness in Visual Recognition: Effective Strategies for Bias Mitigation

Computer vision models learn to perform a task by capturing relevant statistics from training data. It has been shown that models learn spurious age, gender, and race correlations when trained for seemingly unrelated tasks like activity recognition or image captioning. Various mitigation techniques have been presented to prevent models from utilizing or learning such biases. However, there has been little systematic comparison between these techniques. We design a simple but surprisingly effective visual recognition benchmark for studying bias mitigation. Using this benchmark, we provide a thorough analysis of a wide range of techniques. We highlight the shortcomings of popular adversarial training approaches for bias mitigation, propose a simple but similarly effective alternative to the inference-time Reducing Bias Amplification method of Zhao et al., and design a domain-independent training technique that outperforms all other methods. Finally, we validate our findings on the attribute classification task in the CelebA dataset, where attribute presence is known to be correlated with the gender of people in the image, and demonstrate that the proposed technique is effective at mitigating real-world gender bias.

cs.CV↗

arXiv2019

Text-based Editing of Talking-head Video

Editing talking-head video to change the speech content or to remove filler words is challenging. We propose a novel method to edit talking-head video based on its transcript to produce a realistic output video in which the dialogue of the speaker has been modified, while maintaining a seamless audio-visual flow (i.e. no jump cuts). Our method automatically annotates an input talking-head video with phonemes, visemes, 3D face pose and geometry, reflectance, expression and scene illumination per frame. To edit a video, the user has to only edit the transcript, and an optimization strategy then chooses segments of the input corpus as base material. The annotated parameters corresponding to the selected segments are seamlessly stitched together and used to produce an intermediate video representation in which the lower half of the face is rendered with a parametric face model. Finally, a recurrent video generation network transforms this representation to a photorealistic video that matches the edited transcript. We demonstrate a large variety of edits, such as the addition, removal, and alteration of words, as well as convincing language translation and full sentence synthesis.

cs.CV↗

arXiv2019

Learning Shape Templates with Structured Implicit Functions

Template 3D shapes are useful for many tasks in graphics and vision, including fitting observation data, analyzing shape collections, and transferring shape attributes. Because of the variety of geometry and topology of real-world shapes, previous methods generally use a library of hand-made templates. In this paper, we investigate learning a general shape template from data. To allow for widely varying geometry and topology, we choose an implicit surface representation based on composition of local shape elements. While long known to computer graphics, this representation has not yet been explored in the context of machine learning for vision. We show that structured implicit functions are suitable for learning and allow a network to smoothly and simultaneously fit multiple classes of shapes. The learned shape template supports applications such as shape exploration, correspondence, abstraction, interpolation, and semantic segmentation from an RGB image.

cs.CV↗

arXiv2018

Unsupervised Training for 3D Morphable Model Regression

We present a method for training a regression network from image pixels to 3D morphable model coordinates using only unlabeled photographs. The training loss is based on features from a facial recognition network, computed on-the-fly by rendering the predicted faces with a differentiable renderer. To make training from features feasible and avoid network fooling effects, we introduce three objectives: a batch distribution loss that encourages the output distribution to match the distribution of the morphable model, a loopback loss that ensures the network can correctly reinterpret its own output, and a multi-view identity loss that compares the features of the predicted 3D face and the input photograph from multiple viewing angles. We train a regression network using these objectives, a set of unlabeled photographs, and the morphable model itself, and demonstrate state-of-the-art results.

cs.CV↗

arXiv2017

Learning Where to Look: Data-Driven Viewpoint Set Selection for 3D Scenes

The use of rendered images, whether from completely synthetic datasets or from 3D reconstructions, is increasingly prevalent in vision tasks. However, little attention has been given to how the selection of viewpoints affects the performance of rendered training sets. In this paper, we propose a data-driven approach to view set selection. Given a set of example images, we extract statistics describing their contents and generate a set of views matching the distribution of those statistics. Motivated by semantic segmentation tasks, we model the spatial distribution of each semantic object category within an image view volume. We provide a search algorithm that generates a sampling of likely candidate views according to the example distribution, and a set selection algorithm that chooses a subset of the candidates that jointly cover the example distribution. Results of experiments with these algorithms on SUNCG indicate that they are indeed able to produce view distributions similar to an example set from NYUDv2 according to the earth mover's distance. Furthermore, the selected views improve performance on semantic segmentation compared to alternative view selection algorithms.

cs.CV↗

arXiv2015

An Experimental Evaluation of the Best-of-Many Christofides' Algorithm for the Traveling Salesman Problem

Recent papers on approximation algorithms for the traveling salesman problem (TSP) have given a new variant on the well-known Christofides' algorithm for the TSP, called the Best-of-Many Christofides' algorithm. The algorithm involves sampling a spanning tree from the solution the standard LP relaxation of the TSP, subject to the condition that each edge is sampled with probability at most its value in the LP relaxation. One then runs Christofides' algorithm on the tree by computing a minimum-cost matching on the odd-degree vertices in the tree, and shortcutting the resulting Eulerian graph to a tour. In this paper we perform an experimental evaluation of the Best-of-Many Christofides' algorithm to see if there are empirical reasons to believe its performance is better than that of Christofides' algorithm. Furthermore, several different sampling schemes have been proposed; we implement several different schemes to determine which ones might be the most promising for obtaining improved performance guarantees over that of Christofides' algorithm. In our experiments, all of the implemented methods perform significantly better than the Christofides' algorithm; an algorithm that samples from a maximum entropy distribution over spanning trees seems to be particularly good, though there are others that perform almost as well.

cs.DS↗