Reversible image conversion (RIC) aims to build a reversible transformation between specific visual content (e.g., short videos) and an embedding image, where the original content can be restored from the embedding when necessary. This work develops
Invertible Image Conversion Net (IICNet) as a generic solution to various RIC tasks due to its strong capacity and task-independent design. Unlike previous encoder-decoder based methods, IICNet maintains a highly invertible structure based on invertible neural networks (INNs) to better preserve the information during conversion. We use a relation module and a channel squeeze layer to improve the INN nonlinearity to extract cross-image relations and the network flexibility, respectively. Experimental results demonstrate that IICNet outperforms the specifically-designed methods on existing RIC tasks and can generalize well to various newly-explored tasks. With our generic IICNet, we no longer need to hand-engineer task-specific embedding networks for rapidly occurring visual content. Our source codes are available at: https://github.com/felixcheng97/IICNet.
This paper addresses the problem of face video inpainting. Existing video inpainting methods target primarily at natural scenes with repetitive patterns. They do not make use of any prior knowledge of the face to help retrieve correspondences for the
corrupted face. They therefore only achieve sub-optimal results, particularly for faces under large pose and expression variations where face components appear very differently across frames. In this paper, we propose a two-stage deep learning method for face video inpainting. We employ 3DMM as our 3D face prior to transform a face between the image space and the UV (texture) space. In Stage I, we perform face inpainting in the UV space. This helps to largely remove the influence of face poses and expressions and makes the learning task much easier with well aligned face features. We introduce a frame-wise attention module to fully exploit correspondences in neighboring frames to assist the inpainting task. In Stage II, we transform the inpainted face regions back to the image space and perform face video refinement that inpaints any background regions not covered in Stage I and also refines the inpainted face regions. Extensive experiments have been carried out which show our method can significantly outperform methods based merely on 2D information, especially for faces under large pose and expression variations.
We propose a novel approach for embedding novel views in a single JPEG image while preserving the perceptual fidelity of the modified JPEG image and the restored novel views. We adopt the popular novel view synthesis representation of multiplane imag
es (MPIs). Our model first encodes 32 MPI layers (totally 128 channels) into a 3-channel JPEG image that can be decoded for MPIs to render novel views, with an embedding capacity of 1024 bits per pixel. We conducted experiments on public datasets with different novel view synthesis methods, and the results show that the proposed method can restore high-fidelity novel views from a slightly modified JPEG image. Furthermore, our method is robust to JPEG compression, color adjusting, and cropping. Our source code will be publicly available.
Scheduled sampling is widely used to mitigate the exposure bias problem for neural machine translation. Its core motivation is to simulate the inference scene during training by replacing ground-truth tokens with predicted tokens, thus bridging the g
ap between training and inference. However, vanilla scheduled sampling is merely based on training steps and equally treats all decoding steps. Namely, it simulates an inference scene with uniform error rates, which disobeys the real inference scene, where larger decoding steps usually have higher error rates due to error accumulations. To alleviate the above discrepancy, we propose scheduled sampling methods based on decoding steps, increasing the selection chance of predicted tokens with the growth of decoding steps. Consequently, we can more realistically simulate the inference scene during training, thus better bridging the gap between training and inference. Moreover, we investigate scheduled sampling based on both training steps and decoding steps for further improvements. Experimentally, our approaches significantly outperform the Transformer baseline and vanilla scheduled sampling on three large-scale WMT tasks. Additionally, our approaches also generalize well to the text summarization task on two popular benchmarks.
Second harmonic generation (SHG), as one of the most significant c{hi}(2) nonlinear optical processes, plays crucial roles in a broad variety of optical and photonic applications. Designing various delicate schemes to achieve highly efficient SHG has
become a long standing and challenging topic in field of nonlinear optics. Despite numerous success on SHG based on birefringent phase matching and quasi-phase matching, so far, modal phase matching (MPM) for SHG in tightly light-confined structures has still in its infancy. Here, we propose a new scheme to realize highly-efficient SHG via MPM by using a nanophotonic LiNbO3 thin-film waveguide consists of two bonded layers with internally reversed polarizations. In such a dual-layer ridge waveguide based on lithium niobate on insulator, upon optical excitation at 1574.6 nm, we observe SHG at 787.3 nm with ultrahigh conversion efficiency of 5,540% /W/cm/cm experimentally. This work advances our understanding on modal-phase-matched SHG and other quadratic optical nonlinear process, offering additional strategies for development of high-performance nonlinear photonic devices in on-chip platforms.
In contrast to offline working fashions, two research paradigms are devised for online learning: (1) Online Meta Learning (OML) learns good priors over model parameters (or learning to learn) in a sequential setting where tasks are revealed one after
another. Although it provides a sub-linear regret bound, such techniques completely ignore the importance of learning with fairness which is a significant hallmark of human intelligence. (2) Online Fairness-Aware Learning. This setting captures many classification problems for which fairness is a concern. But it aims to attain zero-shot generalization without any task-specific adaptation. This therefore limits the capability of a model to adapt onto newly arrived data. To overcome such issues and bridge the gap, in this paper for the first time we proposed a novel online meta-learning algorithm, namely FFML, which is under the setting of unfairness prevention. The key part of FFML is to learn good priors of an online fair classification models primal and dual parameters that are associated with the models accuracy and fairness, respectively. The problem is formulated in the form of a bi-level convex-concave optimization. Theoretic analysis provides sub-linear upper bounds for loss regret and for violation of cumulative fairness constraints. Our experiments demonstrate the versatility of FFML by applying it to classification on three real-world datasets and show substantial improvements over the best prior work on the tradeoff between fairness and classification accuracy
We present a novel approach to automatic image colorization by imitating the imagination process of human experts. Our imagination module is designed to generate color images that are context-correlated with black-and-white photos. Given a black-and-
white image, our imagination module firstly extracts the context information, which is then used to synthesize colorful and diverse images using a conditional image synthesis network (e.g., semantic image synthesis model). We then design a colorization module to colorize the black-and-white images with the guidance of imagination for photorealistic colorization. Experimental results show that our work produces more colorful and diverse results than state-of-the-art image colorization methods. Our source codes will be publicly available.
Human can easily recognize visual objects with lost information: even losing most details with only contour reserved, e.g. cartoon. However, in terms of visual perception of Deep Neural Networks (DNNs), the ability for recognizing abstract objects (v
isual objects with lost information) is still a challenge. In this work, we investigate this issue from an adversarial viewpoint: will the performance of DNNs decrease even for the images only losing a little information? Towards this end, we propose a novel adversarial attack, named textit{AdvDrop}, which crafts adversarial examples by dropping existing information of images. Previously, most adversarial attacks add extra disturbing information on clean images explicitly. Opposite to previous works, our proposed work explores the adversarial robustness of DNN models in a novel perspective by dropping imperceptible details to craft adversarial examples. We demonstrate the effectiveness of textit{AdvDrop} by extensive experiments, and show that this new type of adversarial examples is more difficult to be defended by current defense systems.
Urban rail transit (URT) system plays a dominating role in many megacities like Beijing and Hong Kong. Due to its important role and complex nature, it is always in great need for public agencies to better understand the performance of the URT system
. This paper focuses on an essential and hard problem to estimate the network-wide link travel time and station waiting time using the automatic fare collection (AFC) data in the URT system, which is beneficial to better understand the system-wide real-time operation state. The emerging data-driven techniques, such as computational graph (CG) models in the machine learning field, provide a new solution for solving this problem. In this study, we first formulate a data-driven estimation optimization framework to estimate the link travel time and station waiting time. Then, we cast the estimation optimization model into a CG framework to solve the optimization problem and obtain the estimation results. The methodology is verified on a synthetic URT network and applied to a real-world URT network using the synthetic and real-world AFC data, respectively. Results show the robustness and effectiveness of the CG-based framework. To the best of our knowledge, this is the first time that the CG is applied to the URT. This study can provide critical insights to better understand the operational state in URT.
While deep reinforcement learning has achieved promising results in challenging decision-making tasks, the main bones of its success --- deep neural networks are mostly black-boxes. A feasible way to gain insight into a black-box model is to distill
it into an interpretable model such as a decision tree, which consists of if-then rules and is easy to grasp and be verified. However, the traditional model distillation is usually a supervised learning task under a stationary data distribution assumption, which is violated in reinforcement learning. Therefore, a typical policy distillation that clones model behaviors with even a small error could bring a data distribution shift, resulting in an unsatisfied distilled policy model with low fidelity or low performance. In this paper, we propose to address this issue by changing the distillation objective from behavior cloning to maximizing an advantage evaluation. The novel distillation objective maximizes an approximated cumulative reward and focuses more on disastrous behaviors in critical states, which controls the data shift effect. We evaluate our method on several Gym tasks, a commercial fight game, and a self-driving car simulator. The empirical results show that the proposed method can preserve a higher cumulative reward than behavior cloning and learn a more consistent policy to the original one. Moreover, by examining the extracted rules from the distilled decision trees, we demonstrate that the proposed method delivers reasonable and robust decisions.
