CVPR 2024 Papers

Noisy correspondence that refers to mismatches in cross-modal data pairs is prevalent on human-annotated or web-crawled datasets. Prior approaches to leverage such data mainly consider the application of uni-modal noisy label learning without amending the impact on both cross-modal and intra-modal geometrical structures in multimodal learning. Actually we find that both structures are effective to discriminate noisy correspondence through structural differences when being well-established. Inspired by this observation we introduce a Geometrical Structure Consistency (GSC) method to infer the true correspondence. Specifically GSC ensures the preservation of geometrical structures within and between modalities allowing for the accurate discrimination of noisy samples based on structural differences. Utilizing these inferred true correspondence labels GSC refines the learning of geometrical structures by filtering out the noisy samples. Experiments across four cross-modal datasets confirm that GSC effectively identifies noisy samples and significantly outperforms the current leading methods. Source code is available at https://github.com/MediaBrain-SJTU/GSC.

Neural Radiance Fields (NeRFs) have shown great potential in novel view synthesis. However they struggle to render sharp images when the data used for training is affected by motion blur. On the other hand event cameras excel in dynamic scenes as they measure brightness changes with microsecond resolution and are thus only marginally affected by blur. Recent methods attempt to enhance NeRF reconstructions under camera motion by fusing frames and events. However they face challenges in recovering accurate color content or constrain the NeRF to a set of predefined camera poses harming reconstruction quality in challenging conditions. This paper proposes a novel formulation addressing these issues by leveraging both model- and learning-based modules. We explicitly model the blur formation process exploiting the event double integral as an additional model-based prior. Additionally we model the event-pixel response using an end-to-end learnable response function allowing our method to adapt to non-idealities in the real event-camera sensor. We show on synthetic and real data that the proposed approach outperforms existing deblur NeRFs that use only frames as well as those that combine frames and events by +6.13dB and +2.48dB respectively.

This paper aims to address a common challenge in deep learning-based image transformation methods such as image enhancement and super-resolution which heavily rely on precisely aligned paired datasets with pixel-level alignments. However creating precisely aligned paired images presents significant challenges and hinders the advancement of methods trained on such data. To overcome this challenge this paper introduces a novel and simple Frequency Distribution Loss (FDL) for computing distribution distance within the frequency domain. Specifically we transform image features into the frequency domain using Discrete Fourier Transformation (DFT). Subsequently frequency components (amplitude and phase) are processed separately to form the FDL loss function. Our method is empirically proven effective as a training constraint due to the thoughtful utilization of global information in the frequency domain. Extensive experimental evaluations focusing on image enhancement and super-resolution tasks demonstrate that FDL outperforms existing misalignment-robust loss functions. Furthermore we explore the potential of our FDL for image style transfer that relies solely on completely misaligned data. Our code is available at: https://github.com/eezkni/FDL

One of the main challenges of multimodal learning is the need to combine heterogeneous modalities (e.g. video audio text). For example video and audio are obtained at much higher rates than text and are roughly aligned in time. They are often not synchronized with text which comes as a global context e.g. a title or a description. Furthermore video and audio inputs are of much larger volumes and grow as the video length increases which naturally requires more compute dedicated to these modalities and makes modeling of long-range dependencies harder. We here decouple the multimodal modeling dividing it into separate autoregressive models processing the inputs according to the characteristics of the modalities. We propose a multimodal model consisting of an autoregressive component for the time-synchronized modalities (audio and video) and an autoregressive component for the context modalities which are not necessarily aligned in time but are still sequential. To address the long-sequences of the video-audio inputs we further partition the video and audio sequences in consecutive snippets and autoregressively process their representations. To that end we propose a Combiner mechanism which models the audio-video information jointly producing compact but expressive representations. This allows us to scale to 512 input video frames without increase in model parameters. Our approach achieves the state-of-the-art on multiple well established multimodal benchmarks. It effectively addresses the high computational demand of media inputs by learning compact representations controlling the sequence length of the audio-video feature representations and modeling their dependencies in time.

Recently 3D Gaussian Splatting has demonstrated impressive novel view synthesis results reaching high fidelity and efficiency. However strong artifacts can be observed when changing the sampling rate e.g. by changing focal length or camera distance. We find that the source for this phenomenon can be attributed to the lack of 3D frequency constraints and the usage of a 2D dilation filter. To address this problem we introduce a 3D smoothing filter to constrains the size of the 3D Gaussian primitives based on the maximal sampling frequency induced by the input views. It eliminates high-frequency artifacts when zooming in. Moreover replacing 2D dilation with a 2D Mip filter which simulates a 2D box filter effectively mitigates aliasing and dilation issues. Our evaluation including scenarios such a training on single-scale images and testing on multiple scales validates the effectiveness of our approach.

This paper focuses on self-supervised monocular depth estimation in dynamic scenes trained on monocular videos. Existing methods jointly estimate pixel-wise depth and motion relying mainly on an image reconstruction loss. Dynamic regions remain a critical challenge for these methods due to the inherent ambiguity in depth and motion estimation resulting in inaccurate depth estimation. This paper proposes a self-supervised training framework exploiting pseudo depth labels for dynamic regions from training data. The key contribution of our framework is to decouple depth estimation for static and dynamic regions of images in the training data. We start with an unsupervised depth estimation approach which provides reliable depth estimates for static regions and motion cues for dynamic regions and allows us to extract moving object information at the instance level. In the next stage we use an object network to estimate the depth of those moving objects assuming rigid motions. Then we propose a new scale alignment module to address the scale ambiguity between estimated depths for static and dynamic regions. We can then use the depth labels generated to train an end-to-end depth estimation network and improve its performance. Extensive experiments on the Cityscapes and KITTI datasets show that our self-training strategy consistently outperforms existing self-/unsupervised depth estimation methods.

We introduce a new family of minimal problems for reconstruction from multiple views. Our primary focus is a novel approach to autocalibration a long-standing problem in computer vision. Traditional approaches to this problem such as those based on Kruppa's equations or the modulus constraint rely explicitly on the knowledge of multiple fundamental matrices or a projective reconstruction. In contrast we consider a novel formulation involving constraints on image points the unknown depths of 3D points and a partially specified calibration matrix K. For 2 and 3 views we present a comprehensive taxonomy of minimal autocalibration problems obtained by relaxing some of these constraints. These problems are organized into classes according to the number of views and any assumed prior knowledge of K. Within each class we determine problems with the fewest---or a relatively small number of---solutions. From this zoo of problems we devise three practical solvers. Experiments with synthetic and real data and interfacing our solvers with COLMAP demonstrate that we achieve superior accuracy compared to state-of-the-art calibration methods. The code is available at https://github.com/andreadalcin/MinimalPerspectiveAutocalibration.

Monocular Depth Estimation (MDE) is a fundamental problem in computer vision with numerous applications. Recently LIDAR-supervised methods have achieved remarkable per-pixel depth accuracy in outdoor scenes. However significant errors are typically found in the proximity of depth discontinuities i.e. depth edges which often hinder the performance of depth-dependent applications that are sensitive to such inaccuracies e.g. novel view synthesis and augmented reality. Since direct supervision for the location of depth edges is typically unavailable in sparse LIDAR-based scenes encouraging the MDE model to produce correct depth edges is not straightforward. To the best of our knowledge this paper is the first attempt to address the depth edges issue for LIDAR-supervised scenes. In this work we propose to learn to detect the location of depth edges from densely-supervised synthetic data and use it to generate supervision for the depth edges in the MDE training. To quantitatively evaluate our approach and due to the lack of depth edges GT in LIDAR-based scenes we manually annotated subsets of the KITTI and the DDAD datasets with depth edges ground truth. We demonstrate significant gains in the accuracy of the depth edges with comparable per-pixel depth accuracy on several challenging datasets. Code and datasets are available at https://github.com/liortalker/MindTheEdge.

Crack segmentation datasets make great efforts to obtain the ground truth crack or non-crack labels as clearly as possible. However it can be observed that ambiguities are still inevitable when considering the marginal non-crack region due to low contrast and heterogeneous texture. To solve this problem we propose a novel clustering-inspired representation learning framework which contains a two-phase strategy for automatic crack segmentation. In the first phase a pre-process is proposed to localize the marginal non-crack region. Then we propose an ambiguity-aware segmentation loss (Aseg Loss) that enables crack segmentation models to capture ambiguities in the above regions via learning segmentation variance which allows us to further localize ambiguous regions. In the second phase to learn the discriminative features of the above regions we propose a clustering-inspired loss (CI Loss) that alters the supervision learning of these regions into an unsupervised clustering manner. We demonstrate that the proposed method could surpass the existing crack segmentation models on various datasets and our constructed CrackSeg5k dataset.

We introduce Mind Artist (MindArt) a novel and efficient neural decoding architecture to snap artistic photographs from our mind in a controllable manner. Recently progress has been made in image reconstruction with non-invasive brain recordings but it's still difficult to generate realistic images with high semantic fidelity due to the scarcity of data annotations. Unlike previous methods this work casts the neural decoding into optimal transport (OT) and representation decoupling problems. Specifically under discrete OT theory we design a graph matching-guided neural representation learning framework to seek the underlying correspondences between conceptual semantics and neural signals which yields a natural and meaningful self-supervisory task. Moreover the proposed MindArt structured with multiple stand-alone modal branches enables the seamless incorporation of semantic representation into any visual style information thus leaving it to have multi-modal reconstruction and training-free semantic editing capabilities. By doing so the reconstructed images of MindArt have phenomenal realism both in terms of semantics and appearance. We compare our MindArt with leading alternatives and achieve SOTA performance in different decoding tasks. Importantly our approach can directly generate a series of stylized "mind snapshots" w/o extra optimizations which may open up more potential applications. Code is available at https://github.com/JxuanC/MindArt.

Deep neural networks (DNNs) are vulnerable to adversarial perturbation where an imperceptible perturbation is added to the image that can fool the DNNs. Diffusion-based adversarial purification uses the diffusion model to generate a clean image against such adversarial attacks. Unfortunately the generative process of the diffusion model is also inevitably affected by adversarial perturbation since the diffusion model is also a deep neural network where its input has adversarial perturbation. In this work we propose MimicDiffusion a new diffusion-based adversarial purification technique that directly approximates the generative process of the diffusion model with the clean image as input. Concretely we analyze the differences between the guided terms using the clean image and the adversarial sample. After that we first implement MimicDiffusion based on Manhattan distance. Then we propose two guidance to purify the adversarial perturbation and approximate the clean diffusion model. Extensive experiments on three image datasets including CIFAR-10 CIFAR-100 and ImageNet with three classifier backbones including WideResNet-70-16 WideResNet-28-10 and ResNet-50 demonstrate that MimicDiffusion significantly performs better than the state-of-the-art baselines. On CIFAR-10 CIFAR-100 and ImageNet it achieves 92.67% 61.35% and 61.53% average robust accuracy which are 18.49% 13.23% and 17.64% higher respectively. The code is available at https://github.com/psky1111/MimicDiffusion.

Volumetric optical microscopy using non-diffracting beams enables rapid imaging of 3D volumes by projecting them axially to 2D images but lacks crucial depth information. Addressing this we introduce MicroDiffusion a pioneering tool facilitating high-quality depth-resolved 3D volume reconstruction from limited 2D projections. While existing Implicit Neural Representation (INR) models often yield incomplete outputs and Denoising Diffusion Probabilistic Models (DDPM) excel at capturing details our method integrates INR's structural coherence with DDPM's fine-detail enhancement capabilities. We pretrain an INR model to transform 2D axially-projected images into a preliminary 3D volume. This pretrained INR acts as a global prior guiding DDPM's generative process through a linear interpolation between INR outputs and noise inputs. This strategy enriches the diffusion process with structured 3D information enhancing detail and reducing noise in localized 2D images.By conditioning the diffusion model on the closest 2D projection MicroDiffusion substantially enhances fidelity in resulting 3D reconstructions surpassing INR and standard DDPM outputs with unparalleled image quality and structural fidelity. Our code and dataset are available athttps://github.com/UCSC-VLAA/MicroDiffusion.

3D visual grounding involves matching natural language descriptions with their corresponding objects in 3D spaces. Existing methods often face challenges with accuracy in object recognition and struggle in interpreting complex linguistic queries particularly with descriptions that involve multiple anchors or are view-dependent. In response we present the MiKASA (Multi-Key-Anchor Scene-Aware) Transformer. Our novel end-to-end trained model integrates a self-attention-based scene-aware object encoder and an original multi-key-anchor technique enhancing object recognition accuracy and the understanding of spatial relationships. Furthermore MiKASA improves the explainability of decision-making facilitating error diagnosis. Our model achieves the highest overall accuracy in the Referit3D challenge for both the Sr3D and Nr3D datasets particularly excelling by a large margin in categories that require viewpoint-dependent descriptions.

Text-to-image diffusion models allow seamless generation of personalized images from scant reference photos. Yet these tools in the wrong hands can fabricate misleading or harmful content endangering individuals. To address this problem existing poisoning-based approaches perturb user images in an imperceptible way to render them "unlearnable" from malicious uses. We identify two limitations of these defending approaches: i) sub-optimal due to the hand-crafted heuristics for solving the intractable bilevel optimization and ii) lack of robustness against simple data transformations like Gaussian filtering. To solve these challenges we propose MetaCloak which solves the bi-level poisoning problem with a meta-learning framework with an additional transformation sampling process to craft transferable and robust perturbation. Specifically we employ a pool of surrogate diffusion models to craft transferable and model-agnostic perturbation. Furthermore by incorporating an additional transformation process we design a simple denoising-error maximization loss that is sufficient for causing transformation-robust semantic distortion and degradation in a personalized generation. Extensive experiments on the VGGFace2 and CelebA-HQ datasets show that MetaCloak outperforms existing approaches. Notably MetaCloak can successfully fool online training services like Replicate in a black-box manner demonstrating the effectiveness of MetaCloak in real-world scenarios.

Category-agnostic pose estimation (CAPE) aims to predict keypoints for arbitrary classes given a few support images annotated with keypoints. Existing methods only rely on the features extracted at support keypoints to predict or refine the keypoints on query image but a few support feature vectors are local and inadequate for CAPE. Considering that human can quickly perceive potential keypoints of arbitrary objects we propose a novel framework for CAPE based on such potential keypoints (named as meta-points). Specifically we maintain learnable embeddings to capture inherent information of various keypoints which interact with image feature maps to produce meta-points without any support. The produced meta-points could serve as meaningful potential keypoints for CAPE. Due to the inevitable gap between inherency and annotation we finally utilize the identities and details offered by support keypoints to assign and refine meta-points to desired keypoints in query image. In addition we propose a progressive deformable point decoder and a slacked regression loss for better prediction and supervision. Our novel framework not only reveals the inherency of keypoints but also outperforms existing methods of CAPE. Comprehensive experiments and in-depth studies on large-scale MP-100 dataset demonstrate the effectiveness of our framework.

DensePose provides a pixel-accurate association of images with 3D mesh coordinates but does not provide a 3D mesh while Human Mesh Reconstruction (HMR) systems have high 2D reprojection error as measured by DensePose localization metrics. In this work we introduce MeshPose to jointly tackle DensePose and HMR. For this we first introduce new losses that allow us to use weak DensePose supervision to accurately localize in 2D a subset of the mesh vertices ('VertexPose'). We then lift these vertices to 3D yielding a low-poly body mesh ('MeshPose'). Our system is trained in an end-to-end manner and is the first HMR method to attain competitive DensePose accuracy while also being lightweight and amenable to efficient inference making it suitable for real-time AR applications.

In this paper we propose a new framework for online 3D scene perception. Conventional 3D scene perception methods are offline i.e. take an already reconstructed 3D scene geometry as input which is not applicable in robotic applications where the input data is streaming RGB-D videos rather than a complete 3D scene reconstructed from pre- collected RGB-D videos. To deal with online 3D scene per- ception tasks where data collection and perception should be performed simultaneously the model should be able to process 3D scenes frame by frame and make use of the temporal information. To this end we propose an adapter-based plug-and-play module for the backbone of 3D scene perception model which constructs memory to cache and aggregate the extracted RGB-D features to empower offline models with temporal learning ability. Specifically we propose a queued memory mechanism to cache the supporting point cloud and image features. Then we devise aggregation modules which directly perform on the memory and pass temporal information to current frame. We further propose 3D-to-2D adapter to enhance image features with strong global context. Our adapters can be easily inserted into mainstream offline architectures of different tasks and significantly boost their performance on online tasks. Extensive experiments on ScanNet and SceneNN datasets demonstrate our approach achieves leading performance on three 3D scene perception tasks compared with state-of-the-art online methods by simply finetuning existing offline models without any model and task-specific designs.

Deep functional maps have emerged in recent years as a prominent learning-based framework for non-rigid shape matching problems. While early methods in this domain only focused on learning in the functional domain the latest techniques have demonstrated that by promoting consistency between functional and pointwise maps leads to significant improvements in accuracy. Unfortunately existing approaches rely heavily on the computation of large dense matrices arising from soft pointwise maps which compromises their efficiency and scalability. To address this limitation we introduce a novel memory-scalable and efficient functional map learning pipeline. By leveraging the specific structure of functional maps we offer the possibility to achieve identical results without ever storing the pointwise map in memory. Furthermore based on the same approach we present a differentiable map refinement layer adapted from an existing axiomatic refinement algorithm. Unlike many functional map learning methods which use this algorithm at a post-processing step ours can be easily used at train time enabling to enforce consistency between the refined and initial versions of the map. Our resulting approach is both simpler more efficient and more numerically stable by avoiding differentiation through a linear system while achieving close to state-of-the-art results in challenging scenarios.

We propose a novel echocardiographical video segmentation model by adapting SAM to medical videos to address some long-standing challenges in ultrasound video segmentation including (1) massive speckle noise and artifacts (2) extremely ambiguous boundaries and (3) large variations of targeting objects across frames. The core technique of our model is a temporal-aware and noise-resilient prompting scheme. Specifically we employ a space-time memory that contains both spatial and temporal information to prompt the segmentation of current frame and thus we call the proposed model as MemSAM. In prompting the memory carrying temporal cues sequentially prompt the video segmentation frame by frame. Meanwhile as the memory prompt propagates high-level features it avoids the issue of misidentification caused by mask propagation and improves representation consistency. To address the challenge of speckle noise we further propose a memory reinforcement mechanism which leverages predicted masks to improve the quality of the memory before storing it. We extensively evaluate our method on two public datasets and demonstrate state-of-the-art performance compared to existing models. Particularly our model achieves comparable performance with fully supervised approaches with limited annotations. Codes are available at https://github.com/dengxl0520/MemSAM.