Normalizing flows have recently demonstrated promising results for low-level vision tasks. For image super-resolution (SR), it learns to predict diverse photo-realistic high-resolution (HR) images from the low-resolution (LR) image rather than learning a deterministic mapping. For image rescaling, it achieves high accuracy by jointly modelling the downscaling and upscaling processes. While existing approaches employ specialized techniques for these two tasks, we set out to unify them in a single formulation. In this paper, we propose the hierarchical conditional flow (HCFlow) as a unified framework for image SR and image rescaling. More specifically, HCFlow learns a bijective mapping between HR and LR image pairs by modelling the distribution of the LR image and the rest high-frequency component simultaneously. In particular, the high-frequency component is conditional on the LR image in a hierarchical manner. To further enhance the performance, other losses such as perceptual loss and GAN loss are combined with the commonly used negative log-likelihood loss in training. Extensive experiments on general image SR, face image SR and image rescaling have demonstrated that the proposed HCFlow achieves state-of-the-art performance in terms of both quantitative metrics and visual quality.
This paper presents a real-time online vision framework to jointly recover an indoor scene's 3D structure and semantic label. Given noisy depth maps, a camera trajectory, and 2D semantic labels at train time, the proposed neural network learns to fuse the depth over frames with suitable semantic labels in the scene space. Our approach exploits the joint volumetric representation of the depth and semantics in the scene feature space to solve this task. For a compelling online fusion of the semantic labels and geometry in real-time, we introduce an efficient vortex pooling block while dropping the routing network in online depth fusion to preserve high-frequency surface details. We show that the context information provided by the semantics of the scene helps the depth fusion network learn noise-resistant features. Not only that, it helps overcome the shortcomings of the current online depth fusion method in dealing with thin object structures, thickening artifacts, and false surfaces. Experimental evaluation on the Replica dataset shows that our approach can perform depth fusion at 37, 10 frames per second with an average reconstruction F-score of 88%, and 91%, respectively, depending on the depth map resolution. Moreover, our model shows an average IoU score of 0.515 on the ScanNet 3D semantic benchmark leaderboard.
Due to the fact that fully supervised semantic segmentation methods require sufficient fully-labeled data to work well and can not generalize to unseen classes, few-shot segmentation has attracted lots of research attention. Previous arts extract features from support and query images, which are processed jointly before making predictions on query images. The whole process is based on convolutional neural networks (CNN), leading to the problem that only local information is used. In this paper, we propose a TRansformer-based Few-shot Semantic segmentation method (TRFS). Specifically, our model consists of two modules: Global Enhancement Module (GEM) and Local Enhancement Module (LEM). GEM adopts transformer blocks to exploit global information, while LEM utilizes conventional convolutions to exploit local information, across query and support features. Both GEM and LEM are complementary, helping to learn better feature representations for segmenting query images. Extensive experiments on PASCAL-5i and COCO datasets show that our approach achieves new state-of-the-art performance, demonstrating its effectiveness.
A set of novel approaches for estimating epistemic uncertainty in deep neural networks with a single forward pass has recently emerged as a valid alternative to Bayesian Neural Networks. On the premise of informative representations, these deterministic uncertainty methods (DUMs) achieve strong performance on detecting out-of-distribution (OOD) data while adding negligible computational costs at inference time. However, it remains unclear whether DUMs are well calibrated and can seamlessly scale to real-world applications - both prerequisites for their practical deployment. To this end, we first provide a taxonomy of DUMs, evaluate their calibration under continuous distributional shifts and their performance on OOD detection for image classification tasks. Then, we extend the most promising approaches to semantic segmentation. We find that, while DUMs scale to realistic vision tasks and perform well on OOD detection, the practicality of current methods is undermined by poor calibration under realistic distributional shifts.
Video segmentation, i.e., partitioning video frames into multiple segments or objects, plays a critical role in a broad range of practical applications, e.g., visual effect assistance in movie, scene understanding in autonomous driving, and virtual background creation in video conferencing, to name a few. Recently, due to the renaissance of connectionism in computer vision, there has been an influx of numerous deep learning based approaches that have been dedicated to video segmentation and delivered compelling performance. In this survey, we comprehensively review two basic lines of research in this area, i.e., generic object segmentation (of unknown categories) in videos and video semantic segmentation, by introducing their respective task settings, background concepts, perceived need, development history, and main challenges. We also provide a detailed overview of representative literature on both methods and datasets. Additionally, we present quantitative performance comparisons of the reviewed methods on benchmark datasets. At last, we point out a set of unsolved open issues in this field, and suggest possible opportunities for further research.
Flow-based generative models have shown excellent ability to explicitly learn the probability density function of data via a sequence of invertible transformations. Yet, modeling long-range dependencies over normalizing flows remains understudied. To fill the gap, in this paper, we introduce two types of invertible attention mechanisms for generative flow models. To be precise, we propose map-based and scaled dot-product attention for unconditional and conditional generative flow models. The key idea is to exploit split-based attention mechanisms to learn the attention weights and input representations on every two splits of flow feature maps. Our method provides invertible attention modules with tractable Jacobian determinants, enabling seamless integration of it at any positions of the flow-based models. The proposed attention mechanism can model the global data dependencies, leading to more comprehensive flow models. Evaluation on multiple generation tasks demonstrates that the introduced attention flow idea results in efficient flow models and compares favorably against the state-of-the-art unconditional and conditional generative flow methods.
Recently normalizing flows (NFs) have demonstrated state-of-the-art performance on modeling 3D point clouds while allowing sampling with arbitrary resolution at inference time. However, these flow-based models still require long training times and large models for representing complicated geometries. This work enhances their representational power by applying mixtures of NFs to point clouds. We show that in this more general framework each component learns to specialize in a particular subregion of an object in a completely unsupervised fashion. By instantiating each mixture component with a comparatively small NF we generate point clouds with improved details compared to single-flow-based models while using fewer parameters and considerably reducing the inference runtime. We further demonstrate that by adding data augmentation, individual mixture components can learn to specialize in a semantically meaningful manner. We evaluate mixtures of NFs on generation, autoencoding and single-view reconstruction based on the ShapeNet dataset.
Video super-resolution (VSR), with the aim to restore a high-resolution video from its corresponding low-resolution version, is a spatial-temporal sequence prediction problem. Recently, Transformer has been gaining popularity due to its parallel computing ability for sequence-to-sequence modeling. Thus, it seems to be straightforward to apply the vision Transformer to solve VSR. However, the typical block design of Transformer with a fully connected self-attention layer and a token-wise feed-forward layer does not fit well for VSR due to the following two reasons. First, the fully connected self-attention layer neglects to exploit the data locality because this layer relies on linear layers to compute attention maps. Second, the token-wise feed-forward layer lacks the feature alignment which is important for VSR since this layer independently processes each of the input token embeddings without any interaction among them. In this paper, we make the first attempt to adapt Transformer for VSR. Specifically, to tackle the first issue, we present a spatial-temporal convolutional self-attention layer with a theoretical understanding to exploit the locality information. For the second issue, we design a bidirectional optical flow-based feed-forward layer to discover the correlations across different video frames and also align features. Extensive experiments on several benchmark datasets demonstrate the effectiveness of our proposed method. The code will be available at https://github.com/caojiezhang/VSR-Transformer.