CVPR 2024 Papers

In this paper we study a new problem Film Removal (FR) which attempts to remove the interference of wrinkled transparent films and reconstruct the original information under films for industrial recognition systems. We first physically model the imaging of industrial materials covered by the film. Considering the specular highlight from the film can be effectively recorded by the polarized camera we build a practical dataset with polarization information containing paired data with and without transparent film. We aim to remove interference from the film (specular highlights and other degradations) with an end-to-end framework. To locate the specular highlight we use an angle estimation network to optimize the polarization angle with the minimized specular highlight. The image with minimized specular highlight is set as a prior for supporting the reconstruction network. Based on the prior and the polarized images the reconstruction network can decouple all degradations from the film. Extensive experiments show that our framework achieves SOTA performance in both image reconstruction and industrial downstream tasks. Our code will be released at https://github.com/jqtangust/FilmRemoval.

Images suffer from heavy spatial redundancy because pixels in neighboring regions are spatially correlated. Existing approaches strive to overcome this limitation by reducing less meaningful image regions. However current leading methods rely on supervisory signals. They may compel models to preserve content that aligns with labeled categories and discard content belonging to unlabeled categories. This categorical inductive bias makes these methods less effective in real-world scenarios. To address this issue we propose a self-supervised framework for image redundancy reduction called Learning to Rank Patches (LTRP). We observe that image reconstruction of masked image modeling models is sensitive to the removal of visible patches when the masking ratio is high (e.g. 90%). Building upon it we implement LTRP via two steps: inferring the semantic density score of each patch by quantifying variation between reconstructions with and without this patch and learning to rank the patches with the pseudo score. The entire process is self-supervised thus getting out of the dilemma of categorical inductive bias. We design extensive experiments on different datasets and tasks. The results demonstrate that LTRP outperforms both supervised and other self-supervised methods due to the fair assessment of image content.

Current metrics for text-to-image models typically rely on statistical metrics which inadequately represent the real preference of humans. Although recent work attempts to learn these preferences via human annotated images they reduce the rich tapestry of human preference to a single overall score. However the preference results vary when humans evaluate images with different aspects. Therefore to learn the multi-dimensional human preferences we propose the Multi-dimensional Preference Score (MPS) the first multi-dimensional preference scoring model for the evaluation of text-to-image models. The MPS introduces the preference condition module upon CLIP model to learn these diverse preferences. It is trained based on our Multi-dimensional Human Preference (MHP) Dataset which comprises 918315 human preference choices across four dimensions (i.e. aesthetics semantic alignment detail quality and overall assessment) on 607541 images. The images are generated by a wide range of latest text-to-image models. The MPS outperforms existing scoring methods across 3 datasets in 4 dimensions enabling it a promising metric for evaluating and improving text-to-image generation. The model and dataset will be made publicly available to facilitate future research.

As the mainstream technique for capturing images of biological specimens at nanometer resolution electron microscopy (EM) is extremely time-consuming for scanning wide field-of-view (FOV) specimens. In this paper we investigate a challenging task of large-factor EM image super-resolution (EMSR) which holds great promise for reducing scanning time relaxing acquisition conditions and expanding imaging FOV. By exploiting the repetitive structures and volumetric coherence of EM images we propose the first generative learning-based framework for large-factor EMSR. Specifically motivated by the predictability of repetitive structures and textures in EM images we first learn a discrete codebook in the latent space to represent high-resolution (HR) cell-specific priors and a latent vector indexer to map low-resolution (LR) EM images to their corresponding latent vectors in a generative manner. By incorporating the generative cell-specific priors from HR EM images through a multi-scale prior fusion module we then deploy multi-image feature alignment and fusion to further exploit the inter-section coherence in the volumetric EM data. Extensive experiments demonstrate that our proposed framework outperforms advanced single-image and video super-resolution methods for 8x and 16x EMSR (i.e. with 64 times and 256 times less data acquired respectively) achieving superior visual reconstruction quality and downstream segmentation accuracy on benchmark EM datasets. Code is available at https://github.com/jtshou/GPEMSR.

This study focuses on a novel task in text-to-image (T2I) generation namely action customization. The objective of this task is to learn the co-existing action from limited data and generalize it to unseen humans or even animals. Experimental results show that existing subject-driven customization methods fail to learn the representative characteristics of actions and struggle in decoupling actions from context features including appearance. To overcome the preference for low-level features and the entanglement of high-level features we propose an inversion-based method Action-Disentangled Identifier (ADI) to learn action-specific identifiers from the exemplar images. ADI first expands the semantic conditioning space by introducing layer-wise identifier tokens thereby increasing the representational richness while distributing the inversion across different features. Then to block the inversion of action-agnostic features ADI extracts the gradient invariance from the constructed sample triples and masks the updates of irrelevant channels. To comprehensively evaluate the task we present an ActionBench that includes a variety of actions each accompanied by meticulously selected samples. Both quantitative and qualitative results show that our ADI outperforms existing baselines in action-customized T2I generation. Our project page is at https://adi-t2i.github.io/ADI.

Diffusion Models have shown remarkable performance in image generation tasks which are capable of generating diverse and realistic image content. When adopting diffusion models for image restoration the crucial challenge lies in how to preserve high-level image fidelity in the randomness diffusion process and generate accurate background structures and realistic texture details. In this paper we propose a general framework and develop a Diffusion Texture Prior Model (DTPM) for image restoration tasks. DTPM explicitly models high-quality texture details through the diffusion process rather than global contextual content. In phase one of the training stage we pre-train DTPM on approximately 55K high-quality image samples after which we freeze most of its parameters. In phase two we insert conditional guidance adapters into DTPM and equip it with an initial predictor thereby facilitating its rapid adaptation to downstream image restoration tasks. Our DTPM could mitigate the randomness of traditional diffusion models by utilizing encapsulated rich and diverse texture knowledge and background structural information provided by the initial predictor during the sampling process. Our comprehensive evaluations of five image restoration tasks demonstrate DTPM's superiority over existing regression and diffusion-based image restoration methods in perceptual quality and its exceptional generalization capabilities.

The difficulty of acquiring high-resolution (HR) and low-resolution (LR) image pairs in real scenarios limits the performance of existing learning-based image super-resolution (SR) methods in the real world. To conduct training on real-world unpaired data current methods focus on synthesizing pseudo LR images to associate unpaired images. However the realness and diversity of pseudo LR images are vulnerable due to the large image space. In this paper we circumvent the difficulty of image generation and propose an alternative to build the connection between unpaired images in a compact proxy space. Specifically we first construct coupled HR and LR dictionaries and then encode HR and LR images into a common latent code space using these dictionaries. In addition we develop an autoencoder-based framework to couple these dictionaries during optimization by reconstructing input HR and LR images. The coupled dictionaries enable our method to employ a shallow network architecture with only 18 layers to achieve efficient image SR. Extensive experiments show that our method (DictSR) can effectively model the LR-to-HR mapping in coupled dictionaries and produces state-of-the-art performance on benchmark datasets.

Current controls over diffusion models (e.g. through text or ControlNet) for image generation fall short in recognizing abstract continuous attributes like illumination direction or non-rigid shape change. In this paper we present an approach for allowing users of text-to-image models to have fine-grained control of several attributes in an image. We do this by engineering special sets of input tokens that can be transformed in a continuous manner we call them Continuous 3D Words. These attributes can for example be represented as sliders and applied jointly with text prompts for fine-grained control over image generation. Given only a single mesh and a rendering engine we show that our approach can be adopted to provide continuous user control over several 3D-aware attributes including time-of-day illumination bird wing orientation dollyzoom effect and object poses. Our method is capable of conditioning image creation with multiple Continuous 3D Words and text descriptions simultaneously while adding no overhead to the generative process.

To synthesize high-fidelity samples diffusion models typically require auxiliary data to guide the generation process. However it is impractical to procure the painstaking patch-level annotation effort required in specialized domains like histopathology and satellite imagery; it is often performed by domain experts and involves hundreds of millions of patches. Modern-day self-supervised learning (SSL) representations encode rich semantic and visual information. In this paper we posit that such representations are expressive enough to act as proxies to fine-grained human labels. We introduce a novel approach that trains diffusion models conditioned on embeddings from SSL. Our diffusion models successfully project these features back to high-quality histopathology and remote sensing images. In addition we construct larger images by assembling spatially consistent patches inferred from SSL embeddings preserving long-range dependencies. Augmenting real data by generating variations of real images improves downstream classifier accuracy for patch-level and larger image-scale classification tasks. Our models are effective even on datasets not encountered during training demonstrating their robustness and generalizability. Generating images from learned embeddings is agnostic to the source of the embeddings. The SSL embeddings used to generate a large image can either be extracted from a reference image or sampled from an auxiliary model conditioned on any related modality (e.g. class labels text genomic data). As proof of concept we introduce the text-to-large image synthesis paradigm where we successfully synthesize large pathology and satellite images out of text descriptions.

Autoregressive Initial Bits (ArIB) a framework that combines subimage autoregression and latent variable models has shown its advantages in lossless image compression. However in current methods the image splitting makes the information of latent variables being uniformly distributed in each subimage and causes inadequate use of latent variables in addition to posterior collapse. To tackle these issues we introduce Bit Plane Slicing (BPS) splitting images in the bit plane dimension with the considerations on different importance for latent variables. Thus BPS provides a more effective representation by arranging subimages with decreasing importance for latent variables. To solve the problem of the increased number of dimensions caused by BPS we further propose a dimension-tailored autoregressive model that tailors autoregression methods for each dimension based on their characteristics efficiently capturing the dependencies in plane space and color dimensions. As shown in the extensive experimental results our method demonstrates the superior compression performance with comparable inference speed when compared to the state-of-the-art normalizing-flow-based methods. The code is at https://github.com/ZZ022/ArIB-BPS.

Recent advances in 3D face stylization have made significant strides in few to zero-shot settings. However the degree of stylization achieved by existing methods is often not sufficient for practical applications because they are mostly based on statistical 3D Morphable Models (3DMM) with limited variations. To this end we propose a method that can produce a highly stylized 3D face model with desired topology. Our methods train a surface deformation network with 3DMM and translate its domain to the target style using a paired exemplar. The network achieves stylization of the 3D face mesh by mimicking the style of the target using a differentiable renderer and directional CLIP losses. Additionally during the inference process we utilize a Mesh Agnostic Encoder (MAGE) that takes deformation target a mesh of diverse topologies as input to the stylization process and encodes its shape into our latent space. The resulting stylized face model can be animated by commonly used 3DMM blend shapes. A set of quantitative and qualitative evaluations demonstrate that our method can produce highly stylized face meshes according to a given style and output them in a desired topology. We also demonstrate example applications of our method including image-based stylized avatar generation linear interpolation of geometric styles and facial animation of stylized avatars.

While diffusion models have significantly advanced the quality of image generation their capability to accurately and coherently render text within these images remains a substantial challenge. Conventional diffusion-based methods for scene text generation are typically limited by their reliance on an intermediate layout output. This dependency often results in a constrained diversity of text styles and fonts an inherent limitation stemming from the deterministic nature of the layout generation phase. To address these challenges this paper introduces SceneTextGen a novel diffusion-based model specifically designed to circumvent the need for a predefined layout stage. By doing so SceneTextGen facilitates a more natural and varied representation of text. The novelty of SceneTextGen lies in its integration of three key components: a character-level encoder for capturing detailed typographic properties coupled with a character-level instance segmentation model and a word-level spotting model to address the issues of unwanted text generation and minor character inaccuracies. We validate the performance of our method by demonstrating improved character recognition rates on generated images across different public visual text datasets in comparison to both standard diffusion based methods and text specific methods.

While replacing Gaussian decoders with a conditional diffusion model enhances the perceptual quality of reconstructions in neural image compression their lack of inductive bias for image data restricts their ability to achieve state-of-the-art perceptual levels. To address this limitation we adopt a non-isotropic diffusion model at the decoder side. This model imposes an inductive bias aimed at distinguishing between frequency contents thereby facilitating the generation of high-quality images. Moreover our framework is equipped with a novel entropy model that accurately models the probability distribution of latent representation by exploiting spatio-channel correlations in latent space while accelerating the entropy decoding step. This channel-wise entropy model leverages both local and global spatial contexts within each channel chunk. The global spatial context is built upon the Transformer which is specifically designed for image compression tasks. The designed Transformer employs a Laplacian-shaped positional encoding the learnable parameters of which are adaptively adjusted for each channel cluster. Our experiments demonstrate that our proposed framework yields better perceptual quality compared to cutting-edge generative-based codecs and the proposed entropy model contributes to notable bitrate savings. The code is available at https://github.com/Atefeh-Khoshtinat/Blur-dissipated-compression.

Composed image retrieval (CIR) task takes a composed query of image and text aiming to search relative images for both conditions. Conventional CIR approaches need a training dataset composed of triplets of query image query text and target image which is very expensive to collect. Several recent works have worked on the zero-shot (ZS) CIR paradigm to tackle the issue without using pre-collected triplets. However the existing ZS-CIR methods show limited backbone scalability and generalizability due to the lack of diversity of the input texts during training. We propose a novel CIR framework only using language for its training. Our LinCIR (Language-only training for CIR) can be trained only with text datasets by a novel self-supervision named self-masking projection (SMP). We project the text latent embedding to the token embedding space and construct a new text by replacing the keyword tokens of the original text. Then we let the new and original texts have the same latent embedding vector. With this simple strategy LinCIR is surprisingly efficient and highly effective; LinCIR with CLIP ViT-G backbone is trained in 48 minutes and shows the best ZS-CIR performances on four different CIR benchmarks CIRCO GeneCIS FashionIQ and CIRR even outperforming supervised method on FashionIQ. Code is available at https://github.com/navervision/lincir

This paper studies the problem of language-guided reflection separation which aims at addressing the ill-posed reflection separation problem by introducing language descriptions to provide layer content. We propose a unified framework to solve this problem which leverages the cross-attention mechanism with contrastive learning strategies to construct the correspondence between language descriptions and image layers. A gated network design and a randomized training strategy are employed to tackle the recognizable layer ambiguity. The effectiveness of the proposed method is validated by the significant performance advantage over existing reflection separation methods on both quantitative and qualitative comparisons.

In this paper we present a novel sequence generation-based framework for lane detection called Lane2Seq. It unifies various lane detection formats by casting lane detection as a sequence generation task. This is different from previous lane detection methods which depend on well-designed task-specific head networks and corresponding loss functions. Lane2Seq only adopts a plain transformer-based encoder-decoder architecture with a simple cross-entropy loss. Additionally we propose a new multi-format model tuning based on reinforcement learning to incorporate the task-specific knowledge into Lane2Seq. Experimental results demonstrate that such a simple sequence generation paradigm not only unifies lane detection but also achieves competitive performance on benchmarks. For example Lane2Seq gets 97.95% and 97.42% F1 score on Tusimple and LLAMAS datasets establishing a new state-of-the-art result for two benchmarks.

The evolution of Diffusion Models has dramatically improved image generation quality making it increasingly difficult to differentiate between real and generated images. This development while impressive also raises significant privacy and security concerns. In response to this we propose a novel Latent REconstruction error guided feature REfinement method (LaRE^2) for detecting the diffusion-generated images. We come up with the Latent Reconstruction Error (LaRE) the first reconstruction-error based feature in the latent space for generated image detection. LaRE surpasses existing methods in terms of feature extraction efficiency while preserving crucial cues required to differentiate between the real and the fake. To exploit LaRE we propose an Error-Guided feature REfinement module (EGRE) which can refine the image feature guided by LaRE to enhance the discriminativeness of the feature. Our EGRE utilizes an align-then-refine mechanism which effectively refines the image feature for generated-image detection from both spatial and channel perspectives. Extensive experiments on the large-scale GenImage benchmark demonstrate the superiority of our LaRE^2 which surpasses the best SoTA method by up to 11.9%/12.1% average ACC/AP across 8 different image generators. LaRE also surpasses existing methods in terms of feature extraction cost delivering an impressive speed enhancement of 8 times.

Weakly-Supervised Scene Graph Generation (WSSGG) research has recently emerged as an alternative to the fully-supervised approach that heavily relies on costly annotations. In this regard studies on WSSGG have utilized image captions to obtain unlocalized triplets while primarily focusing on grounding the unlocalized triplets over image regions. However they have overlooked the two issues involved in the triplet formation process from the captions: 1) Semantic over-simplification issue arises when extracting triplets from captions where fine-grained predicates in captions are undesirably converted into coarse-grained predicates resulting in a long-tailed predicate distribution and 2) Low-density scene graph issue arises when aligning the triplets in the caption with entity/predicate classes of interest where many triplets are discarded and not used in training leading to insufficient supervision. To tackle the two issues we propose a new approach i.e. Large Language Model for weakly-supervised SGG (LLM4SGG) where we mitigate the two issues by leveraging the LLM's in-depth understanding of language and reasoning ability during the extraction of triplets from captions and alignment of entity/predicate classes with target data. To further engage the LLM in these processes we adopt the idea of Chain-of-Thought and the in-context few-shot learning strategy. To validate the effectiveness of LLM4SGG we conduct extensive experiments on Visual Genome and GQA datasets showing significant improvements in both Recall@K and mean Recall@K compared to the state-of-the-art WSSGG methods. A further appeal is that LLM4SGG is data-efficient enabling effective model training with a small amount of training images.

Text-to-image diffusion models have recently received increasing interest for their astonishing ability to produce high-fidelity images from solely text inputs. Subsequent research efforts aim to exploit and apply their capabilities to real image editing. However existing image-to-image methods are often inefficient imprecise and of limited versatility. They either require time-consuming fine-tuning deviate unnecessarily strongly from the input image and/or lack support for multiple simultaneous edits. To address these issues we introduce LEDITS++ an efficient yet versatile and precise textual image manipulation technique. LEDITS++'s novel inversion approach requires no tuning nor optimization and produces high-fidelity results with a few diffusion steps. Second our methodology supports multiple simultaneous edits and is architecture-agnostic. Third we use a novel implicit masking technique that limits changes to relevant image regions. We propose the novel TEdBench++ benchmark as part of our exhaustive evaluation. Our results demonstrate the capabilities of LEDITS++ and its improvements over previous methods.