Learning Foveated Reconstruction to Preserve Perceived Image Statistics

Foveated image reconstruction recovers full image from a sparse set of samples distributed according to the human visual system's retinal sensitivity that rapidly drops with eccentricity. Recently, the use of Generative Adversarial Networks was shown to be a promising solution for such a task as they can successfully hallucinate missing image information. Like for other supervised learning approaches, also for this one, the definition of the loss function and training strategy heavily influences the output quality. In this work, we pose the question of how to efficiently guide the training of foveated reconstruction techniques such that they are fully aware of the human visual system's capabilities and limitations, and therefore, reconstruct visually important image features. Due to the nature of GAN-based solutions, we concentrate on the human's sensitivity to hallucination for different input sample densities. We present new psychophysical experiments, a dataset, and a procedure for training foveated image reconstruction. The strategy provides flexibility to the generator network by penalizing only perceptually important deviations in the output. As a result, the method aims to preserve perceived image statistics rather than natural image statistics. We evaluate our strategy and compare it to alternative solutions using a newly trained objective metric and user experiments.

Towards Coherent Visual Storytelling with Ordered Image Attention

We address the problem of visual storytelling, i.e., generating a story for a given sequence of images. While each sentence of the story should describe a corresponding image, a coherent story also needs to be consistent and relate to both future and past images. To achieve this we develop ordered image attention (OIA). OIA models interactions between the sentence-corresponding image and important regions in other images of the sequence. To highlight the important objects, a message-passing-like algorithm collects representations of those objects in an order-aware manner. To generate the story's sentences, we then highlight important image attention vectors with an Image-Sentence Attention (ISA). Further, to alleviate common linguistic mistakes like repetitiveness, we introduce an adaptive prior. The obtained results improve the METEOR score on the VIST dataset by 1%. In addition, an extensive human study verifies coherency improvements and shows that OIA and ISA generated stories are more focused, shareable, and image-grounded.

MonoPCNS: Monocular 3D Object Detection via Point Cloud Network Simulation

Monocular 3D object detection is a fundamental but very important task to many applications including autonomous driving, robotic grasping and augmented reality. Existing leading methods tend to estimate the depth of the input image first, and detect the 3D object based on point cloud. This routine suffers from the inherent gap between depth estimation and object detection. Besides, the prediction error accumulation would also affect the performance. In this paper, a novel method named MonoPCNS is proposed. The insight behind introducing MonoPCNS is that we propose to simulate the feature learning behavior of a point cloud based detector for monocular detector during the training period. Hence, during inference period, the learned features and prediction would be similar to the point cloud based detector as possible. To achieve it, we propose one scene-level simulation module, one RoI-level simulation module and one response-level simulation module, which are progressively used for the detector's full feature learning and prediction pipeline. We apply our method to the famous M3D-RPN detector and CaDDN detector, conducting extensive experiments on KITTI and Waymo Open dataset. Results show that our method consistently improves the performance of different monocular detectors for a large margin without changing their network architectures. Our method finally achieves state-of-the-art performance.

Treating Point Cloud as Moving Camera Videos: A No-Reference Quality Assessment Metric

Point cloud is one of the most widely used digital representation formats for 3D contents, the visual quality of which may suffer from noise and geometric shift during the production procedure as well as compression and downsampling during the transmission process. To tackle the challenge of point cloud quality assessment (PCQA), many PCQA methods have been proposed to evaluate the visual quality levels of point clouds by assessing the rendered static 2D projections. Although such projection-based PCQA methods achieve competitive performance with the assistance of mature image quality assessment (IQA) methods, they neglect the dynamic quality-aware information, which does not fully match the fact that observers tend to perceive the point clouds through both static and dynamic views. Therefore, in this paper, we treat the point clouds as moving camera videos and explore the way of dealing with PCQA tasks via using video quality assessment (VQA) methods in a no-reference (NR) manner. First, we generate the captured videos by rotating the camera around the point clouds through four circular pathways. Then we extract both spatial and temporal quality-aware features from the selected key frames and the video clips by using trainable 2D-CNN and pre-trained 3D-CNN models respectively. Finally, the visual quality of point clouds is represented by the regressed video quality values. The experimental results reveal that the proposed method is effective for predicting the visual quality levels of the point clouds and even competitive with full-reference (FR) PCQA methods. The ablation studies further verify the rationality of the proposed framework and confirm the contributions made by the quality-aware features extracted from dynamic views.

Applied monocular reconstruction of parametric faces with domain engineering

Many modern online 3D applications and videogames rely on parametric models of human faces for creating believable avatars. However, manual reproduction of someone's facial likeness with a parametric model is difficult and time-consuming. Machine Learning solution for that task is highly desirable but is also challenging. The paper proposes a novel approach to the so-called Face-to-Parameters problem (F2P for short), aiming to reconstruct a parametric face from a single image. The proposed method utilizes synthetic data, domain decomposition, and domain adaptation for addressing multifaceted challenges in solving the F2P. The open-sourced codebase illustrates our key observations and provides means for quantitative evaluation. The presented approach proves practical in an industrial application; it improves accuracy and allows for more efficient models training. The techniques have the potential to extend to other types of parametric models.

Training Process of Unsupervised Learning Architecture for Gravity Spy Dataset

Transient noise appearing in the data from gravitational-wave detectors frequently causes problems, such as instability of the detectors and overlapping or mimicking gravitational-wave signals. Because transient noise is considered to be associated with the environment and instrument, its classification would help to understand its origin and improve the detector's performance. In a previous study, an architecture for classifying transient noise using a time-frequency 2D image (spectrogram) is proposed, which uses unsupervised deep learning combined with variational autoencoder and invariant information clustering. The proposed unsupervised-learning architecture is applied to the Gravity Spy dataset, which consists of Advanced Laser Interferometer Gravitational-Wave Observatory (Advanced LIGO) transient noises with their associated metadata to discuss the potential for online or offline data analysis. In this study, focused on the Gravity Spy dataset, the training process of unsupervised-learning architecture of the previous study is examined and reported.

HairFIT: Pose-Invariant Hairstyle Transfer via Flow-based Hair Alignment and Semantic-Region-Aware Inpainting

Hairstyle transfer is the task of modifying a source hairstyle to a target one. Although recent hairstyle transfer models can reflect the delicate features of hairstyles, they still have two major limitations. First, the existing methods fail to transfer hairstyles when a source and a target image have different poses (e.g., viewing direction or face size), which is prevalent in the real world. Also, the previous models generate unrealistic images when there is a non-trivial amount of regions in the source image occluded by its original hair. When modifying long hair to short hair, shoulders or backgrounds occluded by the long hair need to be inpainted. To address these issues, we propose a novel framework for pose-invariant hairstyle transfer, HairFIT. Our model consists of two stages: 1) flow-based hair alignment and 2) hair synthesis. In the hair alignment stage, we leverage a keypoint-based optical flow estimator to align a target hairstyle with a source pose. Then, we generate a final hairstyle-transferred image in the hair synthesis stage based on Semantic-region-aware Inpainting Mask (SIM) estimator. Our SIM estimator divides the occluded regions in the source image into different semantic regions to reflect their distinct features during the inpainting. To demonstrate the effectiveness of our model, we conduct quantitative and qualitative evaluations using multi-view datasets, K-hairstyle and VoxCeleb. The results indicate that HairFIT achieves a state-of-the-art performance by successfully transferring hairstyles between images of different poses, which has never been achieved before.

Compressive Sensing Based Sparse MIMO Array Optimization for Wideband Near-Field Imaging

In the area of near-field millimeter-wave imaging, the generalized sparse array synthesis (SAS) method is in great demand. The traditional methods usually employ the greedy algorithms, which may have the convergence problem. This paper proposes a convex optimization model for the multiple-input multiple-output (MIMO) array design based on the compressive sensing (CS) approach. We generate a block shaped reference pattern, to be used as an optimizing target. The pattern occupies the entire imaging area of interest in order to involve the effect of each pixel into the optimization model. In MIMO scenarios, we can fix the transmit subarray and synthesize the receive subarray, and vice versa, or doing the synthesis sequentially. The problems associated with focusing, sidelobes suppression, and grating lobes suppression of the synthesized array are examined in details. Numerical and experimental results demonstrate that the synthesized sparse array can offer better image qualities than the sparse arrays with equally spaced or randomly spaced antennas with the same number of antenna elements.

Deep Automatic Natural Image Matting

Automatic image matting (AIM) refers to estimating the soft foreground from an arbitrary natural image without any auxiliary input like trimap, which is useful for image editing. Prior methods try to learn semantic features to aid the matting process while being limited to images with salient opaque foregrounds such as humans and animals. In this paper, we investigate the difficulties when extending them to natural images with salient transparent/meticulous foregrounds or non-salient foregrounds. To address the problem, a novel end-to-end matting network is proposed, which can predict a generalized trimap for any image of the above types as a unified semantic representation. Simultaneously, the learned semantic features guide the matting network to focus on the transition areas via an attention mechanism. We also construct a test set AIM-500 that contains 500 diverse natural images covering all types along with manually labeled alpha mattes, making it feasible to benchmark the generalization ability of AIM models. Results of the experiments demonstrate that our network trained on available composite matting datasets outperforms existing methods both objectively and subjectively. The source code and dataset are available at https://github.com/JizhiziLi/AIM.

Threat Model-Agnostic Adversarial Defense using Diffusion Models

Deep Neural Networks (DNNs) are highly sensitive to imperceptible malicious perturbations, known as adversarial attacks. Following the discovery of this vulnerability in real-world imaging and vision applications, the associated safety concerns have attracted vast research attention, and many defense techniques have been developed. Most of these defense methods rely on adversarial training (AT) -- training the classification network on images perturbed according to a specific threat model, which defines the magnitude of the allowed modification. Although AT leads to promising results, training on a specific threat model fails to generalize to other types of perturbations. A different approach utilizes a preprocessing step to remove the adversarial perturbation from the attacked image. In this work, we follow the latter path and aim to develop a technique that leads to robust classifiers across various realizations of threat models. To this end, we harness the recent advances in stochastic generative modeling, and means to leverage these for sampling from conditional distributions. Our defense relies on an addition of Gaussian i.i.d noise to the attacked image, followed by a pretrained diffusion process -- an architecture that performs a stochastic iterative process over a denoising network, yielding a high perceptual quality denoised outcome. The obtained robustness with this stochastic preprocessing step is validated through extensive experiments on the CIFAR-10 dataset, showing that our method outperforms the leading defense methods under various threat models.