Motion blur is one of the most common degradation artifacts in dynamic scene photography. This paper reviews the NTIRE 2020 Challenge on Image and Video Deblurring. In this challenge, we present the evaluation results from 3 competition tracks as well as the proposed solutions. Track 1 aims to develop single-image deblurring methods focusing on restoration quality. On Track 2, the image deblurring methods are executed on a mobile platform to find the balance of the running speed and the restoration accuracy. Track 3 targets developing video deblurring methods that exploit the temporal relation between input frames. In each competition, there were 163, 135, and 102 registered participants and in the final testing phase, 9, 4, and 7 teams competed. The winning methods demonstrate the state-ofthe-art performance on image and video deblurring tasks.
Land remote sensing analysis is a crucial research in earth science. In this work, we focus on a challenging task of land analysis, i.e., automatic extraction of traffic roads from remote sensing data, which has widespread applications in urban development and expansion estimation. Nevertheless, conventional methods either only utilized the limited information of aerial images, or simply fused multimodal information (e.g., vehicle trajectories), thus cannot well recognize unconstrained roads. To facilitate this problem, we introduce a novel neural network framework termed Cross-Modal Message Propagation Network (CMMPNet), which fully benefits the complementary different modal data (i.e., aerial images and crowdsourced trajectories). Specifically, CMMPNet is composed of two deep Auto-Encoders for modality-specific representation learning and a tailor-designed Dual Enhancement Module for cross-modal representation refinement. In particular, the complementary information of each modality is comprehensively extracted and dynamically propagated to enhance the representation of another modality. Extensive experiments on three real-world benchmarks demonstrate the effectiveness of our CMMPNet for robust road extraction benefiting from blending different modal data, either using image and trajectory data or image and Lidar data. From the experimental results, we observe that the proposed approach outperforms current state-of-the-art methods by large margins.
A small ISO and a small exposure time are usually used to capture an image in the back or low light conditions which results in an image with negligible motion blur and small noise but look dark. In this paper, a single image brightening algorithm is introduced to brighten such an image. The proposed algorithm includes a unique hybrid learning framework to generate two virtual images with large exposure times. The virtual images are first generated via intensity mapping functions (IMFs) which are computed using camera response functions (CRFs) and this is a model-driven approach. Both the virtual images are then enhanced by using a data-driven approach, i.e. a residual convolutional neural network to approach the ground truth images. The model-driven approach and the data-driven one compensate each other in the proposed hybrid learning framework. The final brightened image is obtained by fusing the original image and two virtual images via a multi-scale exposure fusion algorithm with properly defined weights. Experimental results show that the proposed brightening algorithm outperforms existing algorithms in terms of the MEF-SSIM metric.
Image animation transfers the motion of a driving video to a static object in a source image, while keeping the source identity unchanged. Great progress has been made in unsupervised motion transfer recently, where no labelled data or ground truth domain priors are needed. However, current unsupervised approaches still struggle when there are large motion or viewpoint discrepancies between the source and driving images. In this paper, we introduce three measures that we found to be effective for overcoming such large viewpoint changes. Firstly, to achieve more fine-grained motion deformation fields, we propose to apply Neural-ODEs for parametrizing the evolution dynamics of the motion transfer from source to driving. Secondly, to handle occlusions caused by large viewpoint and motion changes, we take advantage of the appearance flow obtained from the source image itself ("self-appearance"), which essentially "borrows" similar structures from other regions of an image to inpaint missing regions. Finally, our framework is also able to leverage the information from additional reference views which help to drive the source identity in spite of varying motion state. Extensive experiments demonstrate that our approach outperforms the state-of-the-arts by a significant margin (~40%), across six benchmarks varying from human faces, human bodies to robots and cartoon characters. Model generality analysis indicates that our approach generalises the best across different object categories as well.
Convolution is an equivariant operation, and image position does not affect its result. A recent study shows that the zero-padding employed in convolutional layers of CNNs provides position information to the CNNs. The study further claims that the position information enables accurate inference for several tasks, such as object recognition, segmentation, etc. However, there is a technical issue with the design of the experiments of the study, and thus the correctness of the claim is yet to be verified. Moreover, the absolute image position may not be essential for the segmentation of natural images, in which target objects will appear at any image position. In this study, we investigate how positional information is and can be utilized for segmentation tasks. Toward this end, we consider {\em positional encoding} (PE) that adds channels embedding image position to the input images and compare PE with several padding methods. Considering the above nature of natural images, we choose medical image segmentation tasks, in which the absolute position appears to be relatively important, as the same organs (of different patients) are captured in similar sizes and positions. We draw a mixed conclusion from the experimental results; the positional encoding certainly works in some cases, but the absolute image position may not be so important for segmentation tasks as we think.
The problem of structure from motion is concerned with recovering the 3-dimensional structure of an object from a set of 2-dimensional images. Generally, all information can be uniquely recovered if enough images and image points are provided, yet there are certain cases where unique recovery is impossible; these are called critical configurations. In this paper we use an algebraic approach to study the critical configurations for three projective cameras. We show that all critical configurations lie on the intersection of quadric surfaces, and classify exactly which intersections constitute a critical configuration.
Localizing objects in image collections without supervision can help to avoid expensive annotation campaigns. We propose a simple approach to this problem, that leverages the activation features of a vision transformer pre-trained in a self-supervised manner. Our method, LOST, does not require any external object proposal nor any exploration of the image collection; it operates on a single image. Yet, we outperform state-of-the-art object discovery methods by up to 8 CorLoc points on PASCAL VOC 2012. We also show that training a class-agnostic detector on the discovered objects boosts results by another 7 points. Moreover, we show promising results on the unsupervised object discovery task. The code to reproduce our results can be found at https://github.com/valeoai/LOST.
Few-shot object detection aims at detecting novel objects with only a few annotated examples. Prior works have proved meta-learning a promising solution, and most of them essentially address detection by meta-learning over regions for their classification and location fine-tuning. However, these methods substantially rely on initially well-located region proposals, which are usually hard to obtain under the few-shot settings. This paper presents a novel meta-detector framework, namely Meta-DETR, which eliminates region-wise prediction and instead meta-learns object localization and classification at image level in a unified and complementary manner. Specifically, it first encodes both support and query images into category-specific features and then feeds them into a category-agnostic decoder to directly generate predictions for specific categories. To facilitate meta-learning with deep networks, we design a simple but effective Semantic Alignment Mechanism (SAM), which aligns high-level and low-level feature semantics to improve the generalization of meta-learned representations. Experiments over multiple few-shot object detection benchmarks show that Meta-DETR outperforms state-of-the-art methods by large margins.
Personality image captioning (PIC) aims to describe an image with a natural language caption given a personality trait. In this work, we introduce a novel formulation for PIC based on a communication game between a speaker and a listener. The speaker attempts to generate natural language captions while the listener encourages the generated captions to contain discriminative information about the input images and personality traits. In this way, we expect that the generated captions can be improved to naturally represent the images and express the traits. In addition, we propose to adapt the language model GPT2 to perform caption generation for PIC. This enables the speaker and listener to benefit from the language encoding capacity of GPT2. Our experiments show that the proposed model achieves the state-of-the-art performance for PIC.
This thesis introduces the wide multiple baseline stereo (WxBS) problem. WxBS, a generalization of the standard wide baseline stereo problem, considers the matching of images that simultaneously differ in more than one image acquisition factor such as viewpoint, illumination, sensor type, or where object appearance changes significantly, e.g., over time. A new dataset with the ground truth, evaluation metric and baselines has been introduced. The thesis presents the following improvements of the WxBS pipeline. (i) A loss function, called HardNeg, for learning a local image descriptor that relies on hard negative mining within a mini-batch and on the maximization of the distance between the closest positive and the closest negative patches. (ii) The descriptor trained with the HardNeg loss, called HardNet, is compact and shows state-of-the-art performance in standard matching, patch verification and retrieval benchmarks. (iii) A method for learning the affine shape, orientation, and potentially other parameters related to geometric and appearance properties of local features. (iv) A tentative correspondences generation strategy which generalizes the standard first to second closest distance ratio is presented. The selection strategy, which shows performance superior to the standard method, is applicable to either hard-engineered descriptors like SIFT, LIOP, and MROGH or deeply learned like HardNet. (v) A feedback loop is introduced for the two-view matching problem, resulting in MODS -- matching with on-demand view synthesis -- algorithm. MODS is an algorithm that handles a viewing angle difference even larger than the previous state-of-the-art ASIFT algorithm, without a significant increase of computational cost over "standard" wide and narrow baseline approaches. Last, but not least, a comprehensive benchmark for local features and robust estimation algorithms is introduced.