Search for: All records

We propose SinGAN-GIF, an extension of the image-based SinGAN [27] to GIFs or short video snippets. Our method learns the distribution of both the image patches in the GIF as well as their motion patterns. We do so by using a pyramid of 3D and 2D convolutional networks to model temporal information while reducing model parameters and training time, along with an image and a video discriminator. SinGAN-GIF can generate similar looking video samples for natural scenes at different spatial resolutions or temporal frame rates, and can be extended to other video applications like video editing, super resolution, and motion transfer. The project page, with supplementary video results, is: https://rajat95.github. io/singan-gif/ 

We present MixNMatch, a conditional generative model that learns to disentangle and encode background, object pose, shape, and texture from real images with minimal supervision, for mix-and-match image generation. We build upon FineGAN, an unconditional generative model, to learn the desired disentanglement and image generator, and leverage adversarial joint image-code distribution matching to learn the latent factor encoders. MixNMatch requires bounding boxes during training to model background, but requires no other supervision. Through extensive experiments, we demonstrate MixNMatch's ability to accurately disentangle, encode, and combine multiple factors for mix-and-match image generation, including sketch2color, cartoon2img, and img2gif applications. Our code/models/demo can be found at https://github.com/Yuheng-Li/MixNMatch 

Card-not-present credit card fraud costs businesses billions of dollars a year. In this paper, we present Boxer, a mobile SDK and server that enables apps to combat card-not-present fraud by scanning cards and verifying that they are genuine. Boxer analyzes the images from these scans, looking for telltale signs of attacks, and introduces a novel abstraction on top of modern security hardware for complementary protection. Currently, 323 apps have integrated Boxer, and tens of them have deployed it to production, including some large, popular, and international apps, resulting in Boxer scanning over 10 million real cards already. Our evaluation of Boxer from one of these deployments shows ten cases of real attacks that our novel hardware-based abstraction detects. Additionally, from the same deployment, without letting in any fraud, Boxer’s card scanning recovers 89% of the good users whom the app would have blocked. In another evaluation of Boxer, we run our image analysis models against images from real users and show an accuracy of 96% and 100% on the two models that we use. 

We propose a novel unsupervised generative model that learns to disentangle object identity from other low-level aspects in class-imbalanced data. We first investigate the issues surrounding the assumptions about uniformity made by InfoGAN [10], and demonstrate its ineffectiveness to properly disentangle object identity in imbalanced data. Our key idea is to make the discovery of the discrete latent factor of variation invariant to identity-preserving transformations in real images, and use that as a signal to learn the appropriate latent distribution representing object identity. Experiments on both artificial (MNIST, 3D cars, 3D chairs, ShapeNet) and real-world (YouTube-Faces) imbalanced datasets demonstrate the effectiveness of our method in disentangling object identity as a latent factor of variation. 

We propose FineGAN, a novel unsupervised GAN framework, which disentangles the background, object shape, and object appearance to hierarchically generate images of fine-grained object categories. To disentangle the factors without supervision, our key idea is to use information theory to associate each factor to a latent code, and to condition the relationships between the codes in a specific way to induce the desired hierarchy. Through extensive experiments, we show that FineGAN achieves the desired disentanglement to generate realistic and diverse images belonging to fine-grained classes of birds, dogs, and cars. Using FineGAN’s automatically learned features, we also cluster real images as a first attempt at solving the novel problem of unsupervised fine-grained object category discovery. Our code/models/demo can be found at https://github.com/kkanshul/finegan 

We propose a novel way of using videos to obtain high precision object proposals for weakly-supervised object detection. Existing weakly-supervised detection approaches use off-the-shelf proposal methods like edge boxes or selective search to obtain candidate boxes. These methods provide high recall but at the expense of thousands of noisy proposals. Thus, the entire burden of finding the few relevant object regions is left to the ensuing object mining step. To mitigate this issue, we focus instead on improving the precision of the initial candidate object proposals. Since we cannot rely on localization annotations, we turn to video and leverage motion cues to automatically estimate the extent of objects to train a Weakly-supervised Region Proposal Network (W-RPN). We use the W-RPN to generate high precision object proposals, which are in turn used to re-rank high recall proposals like edge boxes or selective search according to their spatial overlap. Our W-RPN proposals lead to significant improvement in performance for state-of-the-art weakly-supervised object detection approaches on PASCAL VOC 2007 and 2012. 

We present a novel deep neural network architecture for end-to-end scene flow estimation that directly operates on large-scale 3D point clouds. Inspired by Bilateral Convolutional Layers (BCL), we propose novel DownBCL, UpBCL, and CorrBCL operations that restore structural information from unstructured point clouds, and fuse information from two consecutive point clouds. Operating on discrete and sparse permutohedral lattice points, our architectural design is parsimonious in computational cost. Our model can efficiently process a pair of point cloud frames at once with a maximum of 86K points per frame. Our approach achieves state-of-the-art performance on the FlyingThings3D and KITTI Scene Flow 2015 datasets. Moreover, trained on synthetic data, our approach shows great generalization ability on real-world data and on different point densities without fine-tuning. 

There is an increasing concern in computer vision devices invading the privacy of their users. We want the camera systems/robots to recognize important events and assist human daily life by understanding its videos, but we also want to ensure that they do not intrude people's privacy. In this paper, we propose a new principled approach for learning a video anonymizer. We use an adversarial training setting in which two competing systems fight: (1) a video anonymizer that modifies the original video to remove privacy-sensitive information (i.e., human face) while still trying to maximize spatial action detection performance, and (2) a discriminator that tries to extract privacy-sensitive information from such anonymized videos. The end goal is for the video anonymizer to perform a pixel-level modification of video frames to anonymize each person's face, while minimizing the effect on action detection performance. We experimentally confirm the benefit of our approach particularly compared to conventional hand-crafted video/face anonymization methods including masking, blurring, and noise adding.