ICCV 2023 论文列表

The task of synthesizing novel views from a single image has useful applications in virtual reality and mobile computing, and a number of approaches to the problem have been proposed in recent years. A Multiplane Image (MPI) estimates the scene as a stack of RGBA layers, and can model complex appearance effects, anti-alias depth errors and synthesize soft edges better than methods that use textured meshes or layered depth images. And unlike neural radiance fields, an MPI can be efficiently rendered on graphics hardware. However, MPIs are highly redundant and require a large number of depth layers to achieve plausible results. Based on the observation that the depth complexity in local image regions is lower than that over the entire image, we split an MPI into many small, tiled regions, each with only a few depth planes. We call this representation a Tiled Multiplane Image (TMPI). We propose a method for generating a TMPI with adaptive depth planes for single-view 3D photography in the wild. Our synthesized results are comparable to state-of-the-art single-view MPI methods while having lower computational overhead. each with only a few depth planes. We call this representation a Tiled Multiplane Image (TMPI). We propose a method for generating a TMPI with adaptive depth planes for single-view 3D photography in the wild. Our synthesized results are comparable to state-of-the-art single-view MPI methods while having lower computational overhead.

Point-spread-function (PSF) engineering is a powerful computational imaging technique wherein a custom phase mask is integrated into an optical system to encode additional information into captured images. Used in combination with deep learning, such systems now offer state-of-the-art performance at monocular depth estimation, extended depth-of-field imaging, lensless imaging, and other tasks. Inspired by recent advances in spatial light modulator (SLM) technology, this paper answers a natural question: Can one encode additional information and achieve superior performance by changing a phase mask dynamically over time? We first prove that the set of PSFs described by static phase masks is non-convex and that, as a result, time-averaged PSFs generated by dynamic phase masks are fundamentally more expressive. We then demonstrate, in simulation, that time-averaged dynamic (TiDy) phase masks can leverage this increased expressiveness to offer substantially improved monocular depth estimation and extended depth-of-field imaging performance.

Active learning (AL) aims to select the most useful data samples from an unlabeled data pool and annotate them to expand the labeled dataset under a limited budget.Especially, uncertainty-based methods choose the most uncertain samples, which are known to be effective in improving model performance.However, AL literature often overlooks training dynamics (TD), defined as the ever-changing model behavior during optimization via stochastic gradient descent, even though other research areas have empirically shown that TD provides important clues for measuring the data uncertainty. In this paper, we first provide theoretical and empirical evidence to argue the usefulness of utilizing the ever-changing model behavior rather than the fully trained model snapshot. We then propose a novel AL method, Training Dynamics for Active Learning (TiDAL), which efficiently predicts the training dynamics of unlabeled data to estimate their uncertainty. Experimental results show that our TiDAL achieves better or comparable performance on both balanced and imbalanced benchmark datasets compared to state-of-the-art AL methods, which estimate data uncertainty using only static information after model training.

We present a comprehensive study on a new task named image color aesthetics assessment (ICAA), which aims to assess color aesthetics based on human perception. ICAA is important for various applications such as imaging measurement and image analysis. However, due to the highly diverse aesthetic preferences and numerous color combinations, ICAA presents more challenges than conventional image quality assessment tasks. To advance ICAA research, 1) we propose a baseline model called the Delegate Transformer, which not only deploys deformable transformers to adaptively allocate interest points, but also learns human color space segmentation behavior by the dedicated module. 2) We elaborately build a color-oriented dataset, ICAA17K, containing 17K images, covering 30 popular color combinations, 80 devices and 50 scenes, with each image densely annotated by more than 1,500 people. Moreover, we develop a large-scale benchmark of 15 methods, the most comprehensive one thus far based on two datasets, SPAQ and ICAA17K. Our work, not only achieves state-of-the-art performance, but more importantly offers the community a roadmap to explore solutions for ICAA. Code and dataset are available in https://github.com/woshidandan/Image-Color-Aesthetics-Assessment.

We study the joint image of lines incident to points, meaning the set of image tuples obtained from fixed cameras observing a varying 3D point-line incidence. We prove a formula for the number of complex critical points of the triangulation problem that aims to compute a 3D point-line incidence from noisy images. Our formula works for an arbitrary number of images and measures the intrinsic difficulty of this triangulation. Additionally, we conduct numerical experiments using homotopy continuation methods, comparing different approaches of triangulation of such incidences. In our setup, exploiting the incidence relations gives a notably faster point reconstruction with comparable accuracy.

Source-Free Video Unsupervised Domain Adaptation (SFVUDA) task consists in adapting an action recognition model, trained on a labelled source dataset, to an unlabelled target dataset, without accessing the actual source data. The previous approaches have attempted to address SFVUDA by leveraging self-supervision (e.g., enforcing temporal consistency) derived from the target data itself. In this work, we take an orthogonal approach by exploiting "web-supervision" from Large Language-Vision Models (LLVMs), driven by the rationale that LLVMs contain a rich world prior surprisingly robust to domain-shift. We showcase the unreasonable effectiveness of integrating LLVMs for SFVUDA by devising an intuitive and parameter-efficient method, which we name Domain Adaptation with Large Language-Vision models (DALL-V), that distills the world prior and complementary source model information into a student network tailored for the target. Despite the simplicity, DALL-V achieves significant improvement over state-of-the-art SFVUDA methods.

Generative image modeling enables a wide range of applications but raises ethical concerns about responsible deployment. This paper introduces an active strategy combining image watermarking and Latent Diffusion Models. The goal is for all generated images to conceal a watermark allowing for future detection and/or identification. The method quickly fine-tunes the image generator, conditioned on a binary signature. A pre-trained watermark extractor recovers the hidden signature from any generated image and a statistical test then determines whether it comes from the generative model. We evaluate the invisibility and robustness of our watermark on a variety of generation tasks, showing that Stable Signature works even after the images are modified. For instance, it detects the origin of an image generated from a text prompt, then cropped to keep 10% of the content, with 90+% accuracy at a false positive rate below 1e-6.

In recent years, video generation has become a prominent generative tool and has drawn significant attention. However, there is little consideration in audio-to-video generation, though audio contains unique qualities like temporal semantics and magnitude. Hence, we propose The Power of Sound (TPoS) model to incorporate audio input that includes both changeable temporal semantics and magnitude. To generate video frames, TPoS utilizes a latent stable diffusion model with textual semantic information, which is then guided by the sequential audio embedding from our pretrained Audio Encoder. As a result, this method produces audio reactive video contents. We demonstrate the effectiveness of TPoS across various tasks and compare its results with current state-of-the-art techniques in the field of audio-to-video generation. More examples are available at https://ku-vai.github.io/TPoS/

Semi-supervised learning (SSL) is gaining popularity as it reduces cost of machine learning (ML) by training high performance models using unlabeled data. In this paper, we reveal that the key feature of SSL, i.e., learning from (non-inspected) unlabeled data, exposes SSL to strong poisoning attacks that can significantly damage its security. Poisoning is a long-standing problem in conventional supervised ML, but we argue that, as SSL relies on non-inspected unlabeled data, poisoning poses a more significant threat to SSL. We demonstrate this by designing a backdoor poisoning attack on SSL that can be conducted by a weak adversary with no knowledge of the target SSL pipeline. This is unlike prior poisoning attacks on supervised ML that assume strong adversaries with impractical capabilities. We show that by poisoning only 0.2% of the unlabeled training data, our (weak) adversary can successfully cause misclassification on more than 80% of test inputs (when they contain the backdoor trigger). Our attack remains effective across different benchmark datasets and SSL algorithms, and even circumvents state-of-the-art defenses against backdoor attacks. Our work raises significant concerns about the security of SSL in real-world security critical applications.

Not all camouflages are equally effective, as even a partially visible contour or a slight color difference can make the animal stand out and break its camouflage. In this paper, we address the question of what makes a camouflage successful, by proposing three scores for automatically assessing its effectiveness. In particular, we show that camouflage can be measured by the similarity between background and foreground features and boundary visibility. We use these camouflage scores to assess and compare all available camouflage datasets. We also incorporate the proposed camouflage score into a generative model as an auxiliary loss and show that effective camouflage images or videos can be synthesised in a scalable manner. The generated synthetic dataset is used to train a transformer-based model for segmenting camouflaged animals in videos. Experimentally, we demonstrate state-of-the-art camouflage breaking performance on the public MoCA-Mask benchmark.

Few-shot image generation is a challenging task since it aims to generate diverse new images for an unseen category with only a few images. Existing methods suffer from the trade-off between the quality and diversity of generated images. To tackle this problem, we propose Hyperbolic Attribute Editing (HAE), a simple yet effective method. Unlike prior arts that work in Euclidean space, HAE captures the hierarchy among images using data from seen categories in hyperbolic space. Given a well-trained HAE, images of unseen categories can be generated by moving the latent code of a given image toward any meaningful directions in the Poincare disk with a fixing radius. Most importantly, the hyperbolic space allows us to control the semantic diversity of the generated images by setting different radii in the disk. Extensive experiments and visualizations demonstrate that HAE is capable of not only generating images with promising quality and diversity using limited data but achieving a highly controllable and interpretable editing process.

This paper revisits the standard pretrain-then-finetune paradigm used in computer vision for visual recognition tasks. Typically, state-of-the-art foundation models are pretrained using large scale (weakly) supervised datasets with billions of images. We introduce an additional pre-pretraining stage that is simple and uses the self supervised MAE technique to initialize the model. While MAE has only been shown to scale with the size of models, we find that it scales with the size of the training dataset as well. Thus, our MAE-based pre-pretraining scales with both model and data size making it applicable for training foundation models. Pre-pretraining consistently improves both the model convergence and the downstream transfer performance across a range of model scales (millions to billions of parameters), and dataset sizes (millions to billions of labels). We measure the effectiveness of pre-pretraining on 10 different visual recognition tasks spanning image classification, video recognition, object detection, low-shot classification and zero-shot recognition. Our largest model achieves new state-of-the-art results on iNaturalist-18 (91.3%), 1-shot ImageNet-1k (62.1%), and zero-shot transfer on Food-101 (96.2%). Our study reveals that model initialization plays a significant role, even for web-scale pretraining with billions of images.

Deep Image Prior (DIP) shows that some network architectures inherently tend towards generating smooth images while resisting noise, a phenomenon known as spectral bias. Image denoising is a natural application of this property. Although denoising with DIP mitigates the need for large training sets, two often intertwined practical challenges need to be overcome: architectural design and noise fitting. Existing methods either handcraft or search for suitable architectures from a vast design space, due to the limited understanding of how architectural choices affect the denoising outcome. In this study, we demonstrate from a frequency perspective that unlearnt upsampling is the main driving force behind the denoising phenomenon with DIP. This finding leads to straightforward strategies for identifying a suitable architecture for every image without laborious search. Extensive experiments show that the estimated architectures achieve superior denoising results than existing methods with up to 95% fewer parameters. Thanks to this under-parameterization, the resulting architectures are less prone to noise-fitting.

Cracks are usually curve-like structures that are the focus of many computer-vision applications (e.g., road safety inspection and surface inspection of industrial facilities). The existing pixel-based crack segmentation methods rely on time-consuming and costly pixel-level annotations. And the object-based crack detection methods exploit the horizontal box to detect the crack without considering crack orientation, resulting in scale variation and intra-class variation. Considering this, we provide a new perspective for crack detection that models the cracks as a series of sub-cracks with the corresponding orientation. However, the vanilla adaptation of the existing oriented object detection methods to the crack detection tasks will result in limited performance, due to the boundary discontinuity issue and the ambiguities in sub-crack orientation. In this paper, we propose a first-of-its-kind oriented sub-crack detector, dubbed as CrackDet, which is derived from a novel piecewise angle definition, to ease the boundary discontinuity problem. And then, we propose a multi-branch angle regression loss for learning sub-crack orientation and variance together. Since there are no related benchmarks, we construct three fully annotated datasets, namely, ORC, ONPP, and OCCSD, which involve various cracks in road pavement and industrial facilities. Experiments show that our approach outperforms state-of-the-art crack detectors.

Deep Neural Networks (DNNs)-based semantic segmentation models trained on a source domain often struggle to generalize to unseen target domains, i.e., a domain gap problem. Texture often contributes to the domain gap, making DNNs vulnerable to domain shift because they are prone to be texture-biased. Existing Domain Generalized Semantic Segmentation (DGSS) methods have alleviated the domain gap problem by guiding models to prioritize shape over texture. On the other hand, shape and texture are two prominent and complementary cues in semantic segmentation. This paper argues that leveraging texture is crucial for improving performance in DGSS. Specifically, we propose a novel framework, coined Texture Learning Domain Randomization (TLDR). TLDR includes two novel losses to effectively enhance texture learning in DGSS: (1) a texture regularization loss to prevent overfitting to source domain textures by using texture features from an ImageNet pre-trained model and (2) a texture generalization loss that utilizes random style images to learn diverse texture representations in a self-supervised manner. Extensive experimental results demonstrate the superiority of the proposed TLDR; e.g., TLDR achieves 46.5 mIoU on GTA-to-Cityscapes using ResNet-50, which improves the prior state-of-the-art method by 1.9 mIoU. The source code is available at https://github.com/ssssshwan/TLDR.

Panoptic Scene Graph has recently been proposed for comprehensive scene understanding. However, previous works adopt a fully-supervised learning manner, requiring large amounts of pixel-wise densely-annotated data, which is always tedious and expensive to obtain. To address this limitation, we study a new problem of Panoptic Scene Graph Generation from Purely Textual Descriptions (Caption-to-PSG). The key idea is to leverage the large collection of free image-caption data on the Web alone to generate panoptic scene graphs. The problem is very challenging for three constraints: 1) no location priors; 2) no explicit links between visual regions and textual entities; and 3) no pre-defined concept sets. To tackle this problem, we propose a new framework TextPSG consisting of four modules, i.e., a region grouper, an entity grounder, a segment merger, and a label generator, with several novel techniques. The region grouper first groups image pixels into different segments and the entity grounder then aligns visual segments with language entities based on the textual description of the segment being referred to. The grounding results can thus serve as pseudo labels enabling the segment merger to learn the segment similarity as well as guiding the label generator to learn object semantics and relation predicates, resulting in a fine-grained structured scene understanding. Our framework is effective, significantly outperforming the baselines and achieving strong out-of-distribution robustness. We perform comprehensive ablation studies to corroborate the effectiveness of our design choices and provide an in-depth analysis to highlight future directions. Our code, data, and results are available on our project page: https://vis-www.cs.umass.edu/TextPSG.

We propose TextManiA, a text-driven manifold augmentation method that semantically enriches visual feature spaces, regardless of class distribution. TextManiA augments visual data with intra-class semantic perturbation by exploiting easy-to-understand visually mimetic words, i.e., attributes. This work is built on an interesting hypothesis that general language models, e.g., BERT and GPT, encompass visual information to some extent, even without training on visual training data. Given the hypothesis, TextManiA transfers pre-trained text representation obtained from a well-established large language encoder to a target visual feature space being learned. Our extensive analysis hints that the language encoder indeed encompasses visual information at least useful to augment visual representation. Our experiments demonstrate that TextManiA is particularly powerful in scarce samples with class imbalance as well as even distribution. We also show compatibility with the label mix-based approaches in evenly distributed scarce data.

We present Text2Tex, a novel method for generating high-quality textures for 3D meshes from the given text prompts. Our method incorporates inpainting into a pre-trained depth-aware image diffusion model to progressively synthesize high resolution partial textures from multiple viewpoints. To avoid accumulating inconsistent and stretched artifacts across views, we dynamically segment the rendered view into a generation mask, which represents the generation status of each visible texel. This partitioned view representation guides the depth-aware inpainting model to generate and update partial textures for the corresponding regions. Furthermore, we propose an automatic view sequence generation scheme to determine the next best view for updating the partial texture. Extensive experiments demonstrate that our method significantly outperforms the existing text-driven approaches and GAN-based methods.

We present Text2Room, a method for generating room-scale textured 3D meshes from a given text prompt as input. To this end, we leverage pre-trained 2D text-to-image models to synthesize a sequence of images from different poses. In order to lift these outputs into a consistent 3D scene representation, we combine monocular depth estimation with a text-conditioned inpainting model. The core idea of our approach is a tailored viewpoint selection such that the content of each image can be fused into a seamless, textured 3D mesh. More specifically, we propose a continuous alignment strategy that iteratively fuses scene frames with the existing geometry to create a seamless mesh. Unlike existing works that focus on generating single objects [56, 41] or zoom-out trajectories [18] from text, our method generates complete 3D scenes with multiple objects and explicit 3D geometry. We evaluate our approach using qualitative and quantitative metrics, demonstrating it as the first method to generate room-scale 3D geometry with compelling textures from only text as input.