“Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks”の学習報告

While it is nearly effortless for humans to quickly assess the perceptual similarity between two images, the underlying processes are thought to be quite complex. Despite this, the most widely used perceptual metrics today, such as PSNR and SSIM, are simple, shallow functions, and fail to account for many nuances of human perception. Recently, the deep learning community has found that features of the VGG network trained on ImageNet classification has been remarkably useful as a training loss for image synthesis. But how perceptual are these so-called "perceptual losses"? What elements are critical for their success? To answer these questions, we introduce a new dataset of human perceptual similarity judgments. We systematically evaluate deep features across different architectures and tasks and compare them with classic metrics. We find that deep features outperform all previous metrics by large margins on our dataset. More surprisingly, this result is not restricted to ImageNet-trained VGG features, but holds across different deep architectures and levels of supervision (supervised, self-supervised, or even unsupervised). Our results suggest that perceptual similarity is an emergent property shared across deep visual representations.

The Unreasonable Effectiveness of Deep Features as a Perceptual Metric

Automatic estimation of salient object regions across images, without any prior assumption or knowledge of the contents of the corresponding scenes, enhances many computer vision and computer graphics applications. We introduce a regional contrast based salient object detection algorithm, which simultaneously evaluates global contrast differences and spatial weighted coherence scores. The proposed algorithm is simple, efficient, naturally multi-scale, and produces full-resolution, high-quality saliency maps. These saliency maps are further used to initialize a novel iterative version of GrabCut, namely SaliencyCut, for high quality unsupervised salient object segmentation. We extensively evaluated our algorithm using traditional salient object detection datasets, as well as a more challenging Internet image dataset. Our experimental results demonstrate that our algorithm consistently outperforms 15 existing salient object detection and segmentation methods, yielding higher precision and better recall rates. We also show that our algorithm can be used to efficiently extract salient object masks from Internet images, enabling effective sketch-based image retrieval (SBIR) via simple shape comparisons. Despite such noisy internet images, where the saliency regions are ambiguous, our saliency guided image retrieval achieves a superior retrieval rate compared with state-of-the-art SBIR methods, and additionally provides important target object region information.

https://mmcheng.net/mftp/SalObj/PosterSaliency.pdf

Global contrast based salient region detection

/pdf/the-unreasonable-effectiveness-of-deep-features-as-a-10k66u5qas.pdf

This paper presents interactive image editing tools using a new randomized algorithm for quickly finding approximate nearest-neighbor matches between image patches. Previous research in graphics and vision has leveraged such nearest-neighbor searches to provide a variety of high-level digital image editing tools. However, the cost of computing a field of such matches for an entire image has eluded previous efforts to provide interactive performance. Our algorithm offers substantial performance improvements over the previous state of the art (20-100x), enabling its use in interactive editing tools. The key insights driving the algorithm are that some good patch matches can be found via random sampling, and that natural coherence in the imagery allows us to propagate such matches quickly to surrounding areas. We offer theoretical analysis of the convergence properties of the algorithm, as well as empirical and practical evidence for its high quality and performance. This one simple algorithm forms the basis for a variety of tools -- image retargeting, completion and reshuffling -- that can be used together in the context of a high-level image editing application. Finally, we propose additional intuitive constraints on the synthesis process that offer the user a level of control unavailable in previous methods.

PatchMatch: a randomized correspondence algorithm for structural image editing

Standard video frame interpolation methods first estimate optical flow between input frames and then synthesize an intermediate frame guided by motion. Recent approaches merge these two steps into a single convolution process by convolving input frames with spatially adaptive kernels that account for motion and re-sampling simultaneously. These methods require large kernels to handle large motion, which limits the number of pixels whose kernels can be estimated at once due to the large memory demand. To address this problem, this paper formulates frame interpolation as local separable convolution over input frames using pairs of 1D kernels. Compared to regular 2D kernels, the 1D kernels require significantly fewer parameters to be estimated. Our method develops a deep fully convolutional neural network that takes two input frames and estimates pairs of 1D kernels for all pixels simultaneously. Since our method is able to estimate kernels and synthesizes the whole video frame at once, it allows for the incorporation of perceptual loss to train the neural network to produce visually pleasing frames. This deep neural network is trained end-to-end using widely available video data without any human annotation. Both qualitative and quantitative experiments show that our method provides a practical solution to high-quality video frame interpolation.

/pdf/video-frame-interpolation-via-adaptive-separable-convolution-29lydf1t3k.pdf

Video Frame Interpolation via Adaptive Separable Convolution

We describe a technique that transforms a video from a hand-held video camera so that it appears as if it were taken with a directed camera motion. Our method adjusts the video to appear as if it were taken from nearby viewpoints, allowing 3D camera movements to be simulated. By aiming only for perceptual plausibility, rather than accurate reconstruction, we are able to develop algorithms that can effectively recreate dynamic scenes from a single source video. Our technique first recovers the original 3D camera motion and a sparse set of 3D, static scene points using an off-the-shelf structure-from-motion system. Then, a desired camera path is computed either automatically (e.g., by fitting a linear or quadratic path) or interactively. Finally, our technique performs a least-squares optimization that computes a spatially-varying warp from each input video frame into an output frame. The warp is computed to both follow the sparse displacements suggested by the recovered 3D structure, and avoid deforming the content in the video frame. Our experiments on stabilizing challenging videos of dynamic scenes demonstrate the effectiveness of our technique.

/pdf/content-preserving-warps-for-3d-video-stabilization-42nlys0s4p.pdf

Content-preserving warps for 3D video stabilization

Stereopsis provides an additional depth cue and plays an important role in the human vision system. This paper explores stereopsis for saliency analysis and presents two approaches to stereo saliency detection from stereoscopic images. The first approach computes stereo saliency based on the global disparity contrast in the input image. The second approach leverages domain knowledge in stereoscopic photography. A good stereoscopic image takes care of its disparity distribution to avoid 3D fatigue. Particularly, salient content tends to be positioned in the stereoscopic comfort zone to alleviate the vergence-accommodation conflict. Accordingly, our method computes stereo saliency of an image region based on the distance between its perceived location and the comfort zone. Moreover, we consider objects popping out from the screen salient as these objects tend to catch a viewer's attention. We build a stereo saliency analysis benchmark dataset that contains 1000 stereoscopic images with salient object masks. Our experiments on this dataset show that stereo saliency provides a useful complement to existing visual saliency analysis and our method can successfully detect salient content from images that are difficult for monocular saliency analysis methods.

/pdf/leveraging-stereopsis-for-saliency-analysis-24v5sgk5vn.pdf

Leveraging stereopsis for saliency analysis

Video frame interpolation algorithms typically estimate optical flow or its variations and then use it to guide the synthesis of an intermediate frame between two consecutive original frames. To handle challenges like occlusion, bidirectional flow between the two input frames is often estimated and used to warp and blend the input frames. However, how to effectively blend the two warped frames still remains a challenging problem. This paper presents a context-aware synthesis approach that warps not only the input frames but also their pixel-wise contextual information and uses them to interpolate a high-quality intermediate frame. Specifically, we first use a pre-trained neural network to extract per-pixel contextual information for input frames. We then employ a state-of-the-art optical flow algorithm to estimate bidirectional flow between them and pre-warp both input frames and their context maps. Finally, unlike common approaches that blend the pre-warped frames, our method feeds them and their context maps to a video frame synthesis neural network to produce the interpolated frame in a context-aware fashion. Our neural network is fully convolutional and is trained end to end. Our experiments show that our method can handle challenging scenarios such as occlusion and large motion and outperforms representative state-of-the-art approaches.

/pdf/context-aware-synthesis-for-video-frame-interpolation-sq80914c6t.pdf

Context-Aware Synthesis for Video Frame Interpolation

We present a robust and efficient approach to video stabilization that achieves high-quality camera motion for a wide range of videos. In this article, we focus on the problem of transforming a set of input 2D motion trajectories so that they are both smooth and resemble visually plausible views of the imaged scene; our key insight is that we can achieve this goal by enforcing subspace constraints on feature trajectories while smoothing them. Our approach assembles tracked features in the video into a trajectory matrix, factors it into two low-rank matrices, and performs filtering or curve fitting in a low-dimensional linear space. In order to process long videos, we propose a moving factorization that is both efficient and streamable. Our experiments confirm that our approach can efficiently provide stabilization results comparable with prior 3D methods in cases where those methods succeed, but also provides smooth camera motions in cases where such approaches often fail, such as videos that lack parallax. The presented approach offers the first method that both achieves high-quality video stabilization and is practical enough for consumer applications.

http://web.cecs.pdx.edu/~fliu/papers/tog2010.pdf

Feng Liu

Papers

Video Frame Interpolation via Adaptive Separable Convolution

Content-preserving warps for 3D video stabilization

Leveraging stereopsis for saliency analysis

Context-Aware Synthesis for Video Frame Interpolation

Subspace video stabilization