ICCV 2023 论文列表

LiDAR and cameras are complementary sensors for 3D object detection in autonomous driving. However, it is challenging to explore the unnatural interaction between point clouds and images, and the critical factor is how to conduct feature alignment of heterogeneous modalities. Currently, many methods achieve feature alignment by projection calibration only, without considering the problem of coordinate conversion accuracy errors between sensors, leading to sub-optimal performance. In this paper, we present GraphAlign, a more accurate feature alignment strategy for 3D object detection by graph matching. Specifically, we fuse image features from a semantic segmentation encoder in the image branch and point cloud features from a 3D Sparse CNN in the LiDAR branch. To save computation, we construct the nearest neighbor relationship by calculating Euclidean distance within the subspaces that are divided into the point cloud features. Through the projection calibration between the image and point cloud, we project the nearest neighbors of point cloud features onto the image features. Then by matching the nearest neighbors with a single point cloud to multiple images, we search for a more appropriate feature alignment. In addition, we provide a self-attention module to enhance the weights of significant relations to fine-tune the feature alignment between heterogeneous modalities. Extensive experiments on nuScenes benchmark demonstrate the effectiveness and efficiency of our GraphAlign.

In this paper, we study a novel and widely existing problem in graph matching (GM), namely, Bi-level Noisy Correspondence (BNC), which refers to node-level noisy correspondence (NNC) and edge-level noisy correspondence (ENC). In brief, on the one hand, due to the poor recognizability and viewpoint differences between images, it is inevitable to inaccurately annotate some keypoints with offset and confusion, leading to the mismatch between two associated nodes, i.e., NNC. On the other hand, the noisy node-to-node correspondence will further contaminate the edge-to-edge correspondence, thus leading to ENC. For the BNC challenge, we propose a novel method termed Contrastive Matching with Momentum Distillation. Specifically, the proposed method is with a robust quadratic contrastive loss which enjoys the following merits: i) better exploring the node-to-node and edge-to-edge correlations through a GM customized quadratic contrastive learning paradigm; ii) adaptively penalizing the noisy assignments based on the confidence estimated by the momentum teacher. Extensive experiments on three real-world datasets show the robustness of our model compared with 12 competitive baselines. The code is available at https://github.com/XLearning-SCU/2023-ICCV-COMMON.

We introduce a novel architecture design that enhances expressiveness by incorporating multiple head classifiers (i.e., classification heads) instead of relying on channel expansion or additional building blocks. Our approach employs attention-based aggregation, utilizing pairwise feature similarity to enhance multiple lightweight heads with minimal resource overhead. We compute the Gramian matrices to reinforce class tokens in an attention layer for each head. This enables the heads to learn more discriminative representations, enhancing their aggregation capabilities. Furthermore, we propose a learning algorithm that encourages heads to complement each other by reducing correlation for aggregation. Our models eventually surpass state-of-the-art CNNs and ViTs regarding the accuracy-throughput trade-off on ImageNet-1K and deliver remarkable performance across various downstream tasks, such as COCO object instance segmentation, ADE20k semantic segmentation, and fine-grained visual classification datasets. The effectiveness of our framework is substantiated by practical experimental results and further underpinned by generalization error bound. We release the code publicly at: https://github.com/Lab-LVM/imagenet-models.

Video nystagmography (VNG) is the diagnostic gold standard of benign paroxysmal positional vertigo (BPPV), which requires medical professionals to examine the direction, frequency, intensity, duration, and variation in the strength of nystagmus on a VNG video. This is a tedious process heavily influenced by the doctor's experience, which is error-prone. Recent automatic VNG classification methods approach this problem from the perspective of video analysis without considering medical prior knowledge, resulting in unsatisfactory accuracy and limited diagnostic capability for nystagmographic types, thereby preventing their clinical application. In this paper, we propose an end-to-end data-driven novel BPPV diagnosis framework (TC-BPPV) by considering this problem as an eye trajectory classification problem due to the disease's symptoms and experts' prior knowledge. In this framework, we utilize an eye movement tracking system to capture the eye trajectory and propose the Gram-based attentive neural ordinary differential equations network (Gram-AODE) to perform classification. We validate our framework using the VNG dataset provided by the collaborative university hospital and achieve state-of-the-art performance. We also evaluate Gram-AODE on multiple open-source benchmarks to demonstrate its effectiveness in trajectory classification. Code is available at https://github.com/XiheQiu/Gram-AODE.

Current semi-supervised object detection (SSOD) algorithms typically assume class balanced datasets (PASCAL VOC etc.) or slightly class imbalanced datasets (MSCOCO, etc). This assumption can be easily violated since real world datasets can be extremely class imbalanced in nature, thus making the performance of semi-supervised object detectors far from satisfactory. Besides, the research for this problem in SSOD is severely under-explored. To bridge this research gap, we comprehensively study the class imbalance problem for SSOD under more challenging scenarios, thus forming the first experimental setting for class imbalanced SSOD (CI-SSOD). Moreover, we propose a simple yet effective gradient-based sampling framework that tackles the class imbalance problem from the perspective of two types of confirmation biases. To tackle confirmation bias towards majority classes, the gradient-based reweighting and gradient-based thresholding modules leverage the gradients from each class to fully balance the influence of the majority and minority classes. To tackle the confirmation bias from incorrect pseudo labels of minority classes, the class-rebalancing sampling module resamples unlabeled data following the guidance of the gradient-based reweighting module. Experiments on three proposed sub-tasks, namely MS-COCO, MS-COCO- Object365 and LVIS, suggest that our method outperforms current class imbalanced object detectors by clear margins, serving as a baseline for future research in CISSOD. Code will be available at https://github.com/nightkeepers/CI-SSOD.

Prompt tuning, a recently emerging paradigm, enables the powerful vision-language pre-training models to adapt to downstream tasks in a parameter- and data- efficient way, by learning the "soft prompts" to condition frozen pre-training models. Though effective, it is particularly problematic in the few-shot scenario, where prompt tuning performance is sensitive to the initialization and requires a time-consuming process to find a good initialization, thus restricting the fast adaptation ability of the pre-training models. In addition, prompt tuning could undermine the generalizability of the pre-training models, because the learnable prompt tokens are easy to overfit to the limited training samples. To address these issues, we introduce a novel Gradient-RegulAted Meta-prompt learning (GRAM) framework that jointly meta-learns an efficient soft prompt initialization for better adaptation and a lightweight gradient regulating function for strong cross-domain generalizability in a meta-learning paradigm using only the unlabeled image-text pre-training data. Rather than designing a specific prompt tuning method, our GRAM can be easily incorporated into various prompt tuning methods in a model-agnostic way, and comprehensive experiments show that GRAM brings about consistent improvement for them in several settings (i.e., few-shot learning, cross-domain generalization, cross-dataset generalization, etc.) over 11 datasets. Further, experiments show that GRAM enables the orthogonal methods of textual and visual prompt tuning to work in a mutually-enhanced way, offering better generalizability beyond the uni-modal prompt tuning methods.

Object detection has been expanded from a limited number of categories to open vocabulary. Moving forward, a complete intelligent vision system requires understanding more fine-grained object descriptions, object parts. In this paper, we propose a detector with the ability to predict both open-vocabulary objects and their part segmentation. This ability comes from two designs. First, we train the detector on the joint of part-level, object-level and image-level data to build the multi-granularity alignment between language and image. Second, we parse the novel object into its parts by its dense semantic correspondence with the base object. These two designs enable the detector to largely benefit from various data sources and foundation models. In open-vocabulary part segmentation experiments, our method outperforms the baseline by 3.3 7.3 mAP in cross-dataset generalization on PartImageNet, and improves the baseline by 7.3 novel AP50 in cross-category generalization on Pascal Part. Finally, we train a detector that generalizes to a wide range of part segmentation datasets while achieving better performance than dataset-specific training.

Large-scale pre-trained Vision & Language (VL) models have shown remarkable performance in many applications, enabling replacing a fixed set of supported classes with zero-shot open vocabulary reasoning over (almost arbitrary) natural language prompts. However, recent works have uncovered a fundamental weakness of these models. For example, their difficulty to understand Visual Language Concepts (VLC) that go 'beyond nouns' such as the meaning of non-object words (e.g., attributes, actions, relations, states, etc.), or difficulty in performing compositional reasoning such as understanding the significance of the order of the words in a sentence. In this work, we investigate to which extent purely synthetic data could be leveraged to teach these models to overcome such shortcomings without compromising their zero-shot capabilities. We contribute Synthetic Visual Concepts (SyViC) - a million-scale synthetic dataset and data generation codebase allowing to generate additional suitable data to improve VLC understanding and compositional reasoning of VL models. Additionally, we propose a general VL finetuning strategy for effectively leveraging SyViC towards achieving these improvements. Our extensive experiments and ablations on VL-Checklist, Winoground, and ARO benchmarks demonstrate that it is possible to adapt strong pre-trained VL models with synthetic data significantly enhancing their VLC understanding (e.g. by 9.9% on ARO and 4.3% on VL-Checklist) with under 1% drop in their zero-shot accuracy.

Line segments are powerful features complementary to points. They offer structural cues, robust to drastic viewpoint and illumination changes, and can be present even in texture-less areas. However, describing and matching them is more challenging compared to points due to partial occlusions, lack of texture, or repetitiveness. This paper introduces a new matching paradigm, where points, lines, and their descriptors are unified into a single wireframe structure. We propose GlueStick, a deep matching Graph Neural Network (GNN) that takes two wireframes from different images and leverages the connectivity information between nodes to better glue them together. In addition to the increased efficiency brought by the joint matching, we also demonstrate a large boost of performance when leveraging the complementary nature of these two features in a single architecture. We show that our matching strategy outperforms the state-of-the-art approaches independently matching line segments and points for a wide variety of datasets and tasks. Code is available at https://github.com/cvg/GlueStick.

Text-to-image (T2I) models based on diffusion processes have achieved remarkable success in controllable image generation using user-provided captions. However, the tight coupling between the current text encoder and image decoder in T2I models makes it challenging to replace or upgrade. Such changes often require massive fine-tuning or even training from scratch with the prohibitive expense. To address this problem, we propose GlueGen, which applies a newly proposed GlueNet model to align features from single-modal or multi-modal encoders with the latent space of an existing T2I model. The approach introduces a new training objective that leverages parallel corpora to align the representation spaces of different encoders. Empirical results show that GlueNet can be trained efficiently and enables various capabilities beyond previous state-of-the-art models: 1) multilingual language models such as XLM-Roberta can be aligned with existing T2I models, allowing for the generation of high-quality images from captions beyond English; 2) GlueNet can align multi-modal encoders such as AudioCLIP with the Stable Diffusion model, enabling sound-to-image generation; 3) it can also upgrade the current text encoder of the latent diffusion model for challenging case generation. By the alignment of various feature representations, the GlueNet allows for flexible and efficient integration of new functionality into existing T2I models and sheds light on X-to-image (X2I) generation.

Most in-the-wild images are stored in Low Dynamic Range (LDR) form, serving as a partial observation of the High Dynamic Range (HDR) visual world. Despite limited dynamic range, these LDR images are often captured with different exposures, implicitly containing information about the underlying HDR image distribution. Inspired by this intuition, in this work we present, to the best of our knowledge, the first method for learning a generative model of HDR images from in-the-wild LDR image collections in a fully unsupervised manner. The key idea is to train a generative adversarial network (GAN) to generate HDR images which, when projected to LDR under various exposures, are indistinguishable from real LDR images. Experiments show that our method GlowGAN can synthesize photorealistic HDR images in many challenging cases such as landscapes, lightning, or windows, where previous supervised generative models produce overexposed images. With the assistance of GlowGAN, we showcase the innovative application of unsupervised inverse tone mapping (GlowGAN-ITM) that sets a new paradigm in this field. Unlike previous methods that gradually complete information from LDR input, GlowGAN-ITM searches the entire HDR image manifold modeled by GlowGAN for the HDR images which can be mapped back to the LDR input. GlowGAN-ITM method achieves more realistic reconstruction of overexposed regions compared to state-of-the-art supervised learning models, despite not requiring HDR images or paired multi-exposure images for training.

Sign Language Translation (SLT) is a challenging task due to its cross-domain nature, involving the translation of visual-gestural language to text. Many previous methods employ an intermediate representation,i.e., gloss sequences, to facilitate SLT, thus transforming it into a two-stage task of sign language recognition (SLR) followed by sign language translation (SLT). However, the scarcity of gloss-annotated sign language data, combined with the information bottleneck in the mid-level gloss representation, has hindered the further development of the SLT task. To address this challenge, we propose a novel Gloss-Free SLT base on Visual-Language Pretraining (GFSLT-VLP), which improves SLT by inheriting language-oriented prior knowledge from pre-trained models, without any gloss annotation assistance. Our approach involves two stages: (i) integrating Contrastive Language-Image Pre-training (CLIP) with masked self-supervised learning to create pre-tasks that bridge the semantic gap between visual and textual representations and restore masked sentences, and (ii) constructing an end-to-end architecture with an encoder-decoder-like structure that inherits the parameters of the pre-trained Visual Encoder and Text Decoder from the first stage. The seamless combination of these novel designs forms a robust sign language representation and significantly improves gloss-free sign language translation. In particular, we have achieved unprecedented improvements in terms of BLEU-4 score on the PHOENIX14T dataset (>=+5) and the CSL-Daily dataset (>=+3) compared to state-of-the-art gloss-free SLT methods. Furthermore, our approach also achieves competitive results on the PHOENIX14T dataset when compared with most of the gloss-based methods.

Modeling and designing urban building layouts is of significant interest in computer vision, computer graphics, and urban applications. A building layout consists of a set of buildings in city blocks defined by a network of roads. We observe that building layouts are discrete structures, consisting of multiple rows of buildings of various shapes, and are amenable to skeletonization for mapping arbitrary city block shapes to a canonical form. Hence, we propose a fully automatic approach to building layout generation using a graph attention networks. Our method generates realistic urban layouts given arbitrary road networks, and enables conditional generation based on learned priors. Our results, including user study, demonstrate superior performance as compared to prior layout generation networks, support arbitrary city block and varying building shapes as demonstrated by generating layouts for 28 large cities.

Increasingly, autoregressive approaches are being used to serialize observed variables based on specific criteria. The Neural Processes (NPs) model variable distribution as a continuous function and provide quick solutions for different tasks using a meta-learning framework. This paper proposes an autoregressive-based framework for NPs, based on their autoregressive properties. This framework leverages the autoregressive stacking effects of various variables to enhance the representation of the latent distribution, concurrently refining local and global relationships within the positional representation through the use of a sliding window mechanism. Autoregression improves function approximations in a stacked fashion, thereby raising the upper bound of the optimization. We have designated this framework as Autoregressive Neural Processes (AENPs) or Conditional Autoregressive Neural Processes (CAENPs). Traditional NP models and their variants aim to capture relationships between the context sample points, without addressing either local or global considerations. Specifically, we capture contextual relationships in the deterministic path and introduce sliding window attention and global attention to reconcile local and global relationships in the context sample points. Autoregressive constraints exist between multiple latent variables in the latent paths, thus building a complex global structure that allows our model to learn complex distributions. Finally, we demonstrate the effectiveness of the NPs or CFANPs models for 1D data, Bayesian optimization, and 2D data.

Recent advancements in pre-trained vision-language models, such as CLIP, have enabled the segmentation of arbitrary concepts solely from textual inputs, a process commonly referred to as open-vocabulary semantic segmentation (OVS). However, existing OVS techniques confront a fundamental challenge: the trained classifier tends to overfit on the base classes observed during training, resulting in suboptimal generalization performance to unseen classes. To mitigate this issue, recent studies have proposed the use of an additional frozen pre-trained CLIP for classification. Nonetheless, this approach incurs heavy computational overheads as the CLIP vision encoder must be repeatedly forward-passed for each mask, rendering it impractical for real-world applications. To address this challenge, our objective is to develop a fast OVS model that can perform comparably or better without the extra computational burden of the CLIP image encoder during inference. To this end, we propose a core idea of preserving the generalizable representation when fine-tuning on known classes. Specifically, we introduce a text diversification strategy that generates a set of synonyms for each training category, which prevents the learned representation from collapsing onto specific known category names. Additionally, we employ a text-guided knowledge distillation method to preserve the generalizable knowledge of CLIP. Extensive experiments demonstrate that our proposed model achieves robust generalization performance across various datasets. Furthermore, we perform a preliminary exploration of open-vocabulary video segmentation and present a benchmark that can facilitate future open-vocabulary research in the video domain.

Image retrieval systems conventionally use a two-stage paradigm, leveraging global features for initial retrieval and local features for reranking. However, the scalability of this method is often limited due to the significant storage and computation cost incurred by local feature matching in the reranking stage. In this paper, we present SuperGlobal, a novel approach that exclusively employs global features for both stages, improving efficiency without sacrificing accuracy. SuperGlobal introduces key enhancements to the retrieval system, specifically focusing on the global feature extraction and reranking processes. For extraction, we identify sub-optimal performance when the widely-used ArcFace loss and Generalized Mean (GeM) pooling methods are combined and propose several new modules to improve GeM pooling. In the reranking stage, we introduce a novel method to update the global features of the query and top-ranked images by only considering feature refinement with a small set of images, thus being very compute and memory efficient. Our experiments demonstrate substantial improvements compared to the state of the art in standard benchmarks. Notably, on the Revisited Oxford+1M Hard dataset, our single-stage results improve by 7.1%, while our two-stage gain reaches 3.7% with a strong 64,865x speedup. Our two-stage system surpasses the current single-stage state-of-the-art by 16.3%, offering a scalable, accurate alternative for high-performing image retrieval systems with minimal time overhead. Code: https://github.com/ShihaoShao-GH/SuperGlobal.

Federated learning (FL) is a prevalent distributed machine learning approach that enables collaborative training of a global model across multiple devices without sharing local data. However, the presence of long-tailed data can negatively deteriorate the model's performance in real-world FL applications. Moreover, existing re-balance strategies are less effective for the federated long-tailed issue when directly utilizing local label distribution as the class prior at the clients' side. To this end, we propose a novel Global Balanced Multi-Expert (GBME) framework to optimize a balanced global objective, which does not require additional information beyond the standard FL pipeline. In particular, a proxy is derived from the accumulated gradients uploaded by the clients after local training, and is shared by all clients as the class prior for re-balance training. Such a proxy can also guide the client grouping to train a multi-expert model, where the knowledge from different clients can be aggregated via the ensemble of different experts corresponding to different client groups. To further strengthen the privacy-preserving ability, we present a GBME-p algorithm with a theoretical guarantee to prevent privacy leakage from the proxy. Extensive experiments on long-tailed decentralized datasets demonstrate the effectiveness of GBME and GBME-p, both of which show superior performance to state-of-the-art methods.

When applying a pre-trained 2D-to-3D Human Pose lifting model to a target unseen dataset, a large performance degradation is commonly encountered due to domain shift issues. We observe that the degradation is caused by two factors: 1) the large distribution gap over global positions of poses between the source and target datasets due to variant camera parameters and settings, and 2) the deficient diversity of local structures of poses in the training. To this end, we combine global adaptation and local generalization in PoseDA, a simple yet effective framework of unsupervised domain adaptation for 3D human pose estimation. Specifically, global adaptation aims to align global positions of poses from source domain to target domain with a proposed global position alignment (GPA) module. This module brings significant performance improvement without introducing additional learnable parameters. In addition, we propose local pose augmentation (LPA) to enhance the diversity of 3D poses following an adversarial training scheme consisting of 1) a augmentation generator that generates the parameters of pre-defined pose transformations and 2) an anchor discriminator to ensure the reality and quality of the augmented data. Our approach can be applicable to almost all 2D-3D lifting models. PoseDA achieves 61.3 mm of MPJPE on MPI-INF-3DHP under a cross-dataset evaluation setup, improving upon the previous state-of-the-art method by 10.2%.

Fast generation of high-quality 3D digital humans is important to a vast number of applications ranging from entertainment to professional concerns. Recent advances in differentiable rendering have enabled the training of 3D generative models without requiring 3D ground truths. However, the quality of the generated 3D humans still has much room to improve in terms of both fidelity and diversity. In this paper, we present Get3DHuman, a novel 3D human framework that can significantly boost the realism and diversity of the generated outcomes by only using a limited budget of 3D ground-truth data. Our key observation is that the 3D generator can profit from human-related priors learned through 2D human generators and 3D reconstructors. Specifically, we bridge the latent space of Get3DHuman with that of StyleGAN-Human via a specially-designed prior network, where the input latent code is mapped to the shape and texture feature volumes spanned by the pixel-aligned 3D reconstructor. The outcomes of the prior network are then leveraged as the supervisory signals for the main generator network. To ensure effective training, we further propose three tailored losses applied to the generated feature volumes and the intermediate feature maps. Extensive experiments demonstrate that Get3DHuman greatly outperforms the other state-of-the-art approaches and can support a wide range of applications including shape interpolation, shape re-texturing, and single-view reconstruction through latent inversion.