arXiv subjects

Hongkai Xiong

Publications and source records attributed to Hongkai Xiong.

At least 37 records · Page 2Linked to original sources

arXiv2025

On Disentangled Training for Nonlinear Transform in Learned Image Compression

Learned image compression (LIC) has demonstrated superior rate-distortion (R-D) performance compared to traditional codecs, but is challenged by training inefficiency that could incur more than two weeks to train a state-of-the-art model from scratch. Existing LIC methods overlook the slow convergence caused by compacting energy in learning nonlinear transforms. In this paper, we first reveal that such energy compaction consists of two components, i.e., feature decorrelation and uneven energy modulation. On such basis, we propose a linear auxiliary transform (AuxT) to disentangle energy compaction in training nonlinear transforms. The proposed AuxT obtains coarse approximation to achieve efficient energy compaction such that distribution fitting with the nonlinear transforms can be simplified to fine details. We then develop wavelet-based linear shortcuts (WLSs) for AuxT that leverages wavelet-based downsampling and orthogonal linear projection for feature decorrelation and subband-aware scaling for

eess.IV↗

arXiv2024

Frequency-Aware Transformer for Learned Image Compression

Learned image compression (LIC) has gained traction as an effective solution for image storage and transmission in recent years. However, existing LIC methods are redundant in latent representation due to limitations in capturing anisotropic frequency components and preserving directional details. To overcome these challenges, we propose a novel frequency-aware transformer (FAT) block that for the first time achieves multiscale directional ananlysis for LIC. The FAT block comprises frequency-decomposition window attention (FDWA) modules to capture multiscale and directional frequency components of natural images. Additionally, we introduce frequency-modulation feed-forward network (FMFFN) to adaptively modulate different frequency components, improving rate-distortion performance. Furthermore, we present a transformer-based channel-wise autoregressive (T-CA) model that effectively exploits channel dependencies. Experiments show that our method achieves state-of-the-art rate-distortion performance compared to existing LIC methods, and evidently outperforms latest standardized codec VTM-12.1 by 14.5%, 15.1%, 13.0% in BD-rate on the Kodak, Tecnick, and CLIC datasets.

eess.IV↗

arXiv2024

LiftImage3D: Lifting Any Single Image to 3D Gaussians with Video Generation Priors

Single-image 3D reconstruction remains a fundamental challenge in computer vision due to inherent geometric ambiguities and limited viewpoint information. Recent advances in Latent Video Diffusion Models (LVDMs) offer promising 3D priors learned from large-scale video data. However, leveraging these priors effectively faces three key challenges: (1) degradation in quality across large camera motions, (2) difficulties in achieving precise camera control, and (3) geometric distortions inherent to the diffusion process that damage 3D consistency. We address these challenges by proposing LiftImage3D, a framework that effectively releases LVDMs' generative priors while ensuring 3D consistency. Specifically, we design an articulated trajectory strategy to generate video frames, which decomposes video sequences with large camera motions into ones with controllable small motions. Then we use robust neural matching models, i.e. MASt3R, to calibrate the camera poses of generated frames and produce corresponding point clouds. Finally, we propose a distortion-aware 3D Gaussian splatting representation, which can learn independent distortions between frames and output undistorted canonical Gaussians. Extensive experiments demonstrate that LiftImage3D achieves state-of-the-art performance on two challenging datasets, i.e. LLFF, DL3DV, and Tanks and Temples, and generalizes well to diverse in-the-wild images, from cartoon illustrations to complex real-world scenes.

cs.CV↗

arXiv2024

Point Cloud Resampling with Learnable Heat Diffusion

Generative diffusion models have shown empirical successes in point cloud resampling, generating a denser and more uniform distribution of points from sparse or noisy 3D point clouds by progressively refining noise into structure. However, existing diffusion models employ manually predefined schemes, which often fail to recover the underlying point cloud structure due to the rigid and disruptive nature of the geometric degradation. To address this issue, we propose a novel learnable heat diffusion framework for point cloud resampling, which directly parameterizes the marginal distribution for the forward process by learning the adaptive heat diffusion schedules and local filtering scales of the time-varying heat kernel, and consequently, generates an adaptive conditional prior for the reverse process. Unlike previous diffusion models with a fixed prior, the adaptive conditional prior selectively preserves geometric features of the point cloud by minimizing a refined variational lower bound, guiding the points to evolve towards the underlying surface during the reverse process. Extensive experimental results demonstrate that the proposed point cloud resampling achieves state-of-the-art performance in representative reconstruction tasks including point cloud denoising and upsampling.

cs.CV↗

arXiv2024

Point Cloud Denoising With Fine-Granularity Dynamic Graph Convolutional Networks

Due to limitations in acquisition equipment, noise perturbations often corrupt 3-D point clouds, hindering down-stream tasks such as surface reconstruction, rendering, and further processing. Existing 3-D point cloud denoising methods typically fail to reliably fit the underlying continuous surface, resulting in a degradation of reconstruction performance. This paper introduces fine-granularity dynamic graph convolutional networks called GD-GCN, a novel approach to denoising in 3-D point clouds. The GD-GCN employs micro-step temporal graph convolution (MST-GConv) to perform feature learning in a gradual manner. Compared with the conventional GCN, which commonly uses discrete integer-step graph convolution, this modification introduces a more adaptable and nuanced approach to feature learning within graph convolution networks. It more accurately depicts the process of fitting the point cloud with noise to the underlying surface by and the learning process for MST-GConv acts like a changing system and is managed through a type of neural network known as neural Partial Differential Equations (PDEs). This means it can adapt and improve over time. GD-GCN approximates the Riemannian metric, calculating distances between points along a low-dimensional manifold. This capability allows it to understand the local geometric structure and effectively capture diverse relationships between points from different geometric regions through geometric graph construction based on Riemannian distances. Additionally, GD-GCN incorporates robust graph spectral filters based on the Bernstein polynomial approximation, which modulate eigenvalues for complex and arbitrary spectral responses, providing theoretical guarantees for BIBO stability. Symmetric channel mixing matrices further enhance filter flexibility by enabling channel-level scaling and shifting in the spectral domain.

cs.CV↗

arXiv2024

UMG-CLIP: A Unified Multi-Granularity Vision Generalist for Open-World Understanding

Vision-language foundation models, represented by Contrastive Language-Image Pre-training (CLIP), have gained increasing attention for jointly understanding both vision and textual tasks. However, existing approaches primarily focus on training models to match global image representations with textual descriptions, thereby overlooking the critical alignment between local regions and corresponding text tokens. This paper extends CLIP with multi-granularity alignment. Notably, we deliberately construct a new dataset comprising pseudo annotations at various levels of granularities, encompassing image-level, region-level as well as pixel-level captions and tags. Accordingly, we develop a Unified Multi-Granularity learning framework, termed UMG-CLIP, which simultaneously empowers the model with versatile perception abilities across different levels of detail. With parameter efficient tuning, UMG-CLIP surpasses current widely used CLIP variants and achieves state-of-the-art performance on diverse image understanding benchmarks, including open-world recognition, retrieval, semantic segmentation, and panoptic segmentation tasks. We believe that UMG-CLIP represents a valuable advancement in vision-language foundation models. The code is available at https://github.com/lygsbw/UMG-CLIP.

cs.CV↗

arXiv2024

Cascade-Zero123: One Image to Highly Consistent 3D with Self-Prompted Nearby Views

Synthesizing multi-view 3D from one single image is a significant but challenging task. Zero-1-to-3 methods have achieved great success by lifting a 2D latent diffusion model to the 3D scope. The target view image is generated with a single-view source image and the camera pose as condition information. However, due to the high sparsity of the single input image, Zero-1-to-3 tends to produce geometry and appearance inconsistency across views, especially for complex objects. To tackle this issue, we propose to supply more condition information for the generation model but in a self-prompt way. A cascade framework is constructed with two Zero-1-to-3 models, named Cascade-Zero123, which progressively extract 3D information from the source image. Specifically, several nearby views are first generated by the first model and then fed into the second-stage model along with the source image as generation conditions. With amplified self-prompted condition images, our Cascade-Zero123 generates more consistent novel-view images than Zero-1-to-3. Experiment results demonstrate remarkable promotion, especially for various complex and challenging scenes, involving insects, humans, transparent objects, and stacked multiple objects etc. More demos and code are available at https://cascadezero123.github.io.

cs.CV↗

arXiv2024

Image Compression for Machine and Human Vision with Spatial-Frequency Adaptation

Image compression for machine and human vision (ICMH) has gained increasing attention in recent years. Existing ICMH methods are limited by high training and storage overheads due to heavy design of task-specific networks. To address this issue, in this paper, we develop a novel lightweight adapter-based tuning framework for ICMH, named Adapt-ICMH, that better balances task performance and bitrates with reduced overheads. We propose a spatial-frequency modulation adapter (SFMA) that simultaneously eliminates non-semantic redundancy with a spatial modulation adapter, and enhances task-relevant frequency components and suppresses task-irrelevant frequency components with a frequency modulation adapter. The proposed adapter is plug-and-play and compatible with almost all existing learned image compression models without compromising the performance of pre-trained models. Experiments demonstrate that Adapt-ICMH consistently outperforms existing ICMH frameworks on various machine vision tasks with fewer fine-tuned parameters and reduced computational complexity. Code will be released at https://github.com/qingshi9974/ECCV2024-AdpatICMH .

cs.CV↗

arXiv2024

Betrayed by Attention: A Simple yet Effective Approach for Self-supervised Video Object Segmentation

In this paper, we propose a simple yet effective approach for self-supervised video object segmentation (VOS). Our key insight is that the inherent structural dependencies present in DINO-pretrained Transformers can be leveraged to establish robust spatio-temporal correspondences in videos. Furthermore, simple clustering on this correspondence cue is sufficient to yield competitive segmentation results. Previous self-supervised VOS techniques majorly resort to auxiliary modalities or utilize iterative slot attention to assist in object discovery, which restricts their general applicability and imposes higher computational requirements. To deal with these challenges, we develop a simplified architecture that capitalizes on the emerging objectness from DINO-pretrained Transformers, bypassing the need for additional modalities or slot attention. Specifically, we first introduce a single spatio-temporal Transformer block to process the frame-wise DINO features and establish spatio-temporal dependencies in the form of self-attention. Subsequently, utilizing these attention maps, we implement hierarchical clustering to generate object segmentation masks. To train the spatio-temporal block in a fully self-supervised manner, we employ semantic and dynamic motion consistency coupled with entropy normalization. Our method demonstrates state-of-the-art performance across multiple unsupervised VOS benchmarks and particularly excels in complex real-world multi-object video segmentation tasks such as DAVIS-17-Unsupervised and YouTube-VIS-19. The code and model checkpoints will be released at https://github.com/shvdiwnkozbw/SSL-UVOS.

cs.CV↗

arXiv2024

Improving Diffusion Models for Inverse Problems Using Optimal Posterior Covariance

Recent diffusion models provide a promising zero-shot solution to noisy linear inverse problems without retraining for specific inverse problems. In this paper, we reveal that recent methods can be uniformly interpreted as employing a Gaussian approximation with hand-crafted isotropic covariance for the intractable denoising posterior to approximate the conditional posterior mean. Inspired by this finding, we propose to improve recent methods by using more principled covariance determined by maximum likelihood estimation. To achieve posterior covariance optimization without retraining, we provide general plug-and-play solutions based on two approaches specifically designed for leveraging pre-trained models with and without reverse covariance. We further propose a scalable method for learning posterior covariance prediction based on representation with orthonormal basis. Experimental results demonstrate that the proposed methods significantly enhance reconstruction performance without requiring hyperparameter tuning.

cs.CV↗

arXiv2024

CPPF++: Uncertainty-Aware Sim2Real Object Pose Estimation by Vote Aggregation

Object pose estimation constitutes a critical area within the domain of 3D vision. While contemporary state-of-the-art methods that leverage real-world pose annotations have demonstrated commendable performance, the procurement of such real training data incurs substantial costs. This paper focuses on a specific setting wherein only 3D CAD models are utilized as a priori knowledge, devoid of any background or clutter information. We introduce a novel method, CPPF++, designed for sim-to-real pose estimation. This method builds upon the foundational point-pair voting scheme of CPPF, reformulating it through a probabilistic view. To address the challenge posed by vote collision, we propose a novel approach that involves modeling the voting uncertainty by estimating the probabilistic distribution of each point pair within the canonical space. Furthermore, we augment the contextual information provided by each voting unit through the introduction of N-point tuples. To enhance the robustness and accuracy of the model, we incorporate several innovative modules, including noisy pair filtering, online alignment optimization, and a tuple feature ensemble. Alongside these methodological advancements, we introduce a new category-level pose estimation dataset, named DiversePose 300. Empirical evidence demonstrates that our method significantly surpasses previous sim-to-real approaches and achieves comparable or superior performance on novel datasets. Our code is available on https://github.com/qq456cvb/CPPF2.

cs.CV↗

arXiv2024

From CLIP to DINO: Visual Encoders Shout in Multi-modal Large Language Models

Multi-modal Large Language Models (MLLMs) have made significant strides in expanding the capabilities of Large Language Models (LLMs) through the incorporation of visual perception interfaces. Despite the emergence of exciting applications and the availability of diverse instruction tuning data, existing approaches often rely on CLIP or its variants as the visual branch, and merely extract features from the deep layers. However, these methods lack a comprehensive analysis of the visual encoders in MLLMs. In this paper, we conduct an extensive investigation into the effectiveness of different vision encoders within MLLMs. Our findings reveal that the shallow layer features of CLIP offer particular advantages for fine-grained tasks such as grounding and region understanding. Surprisingly, the vision-only model DINO, which is not pretrained with text-image alignment, demonstrates promising performance as a visual branch within MLLMs. By simply equipping it with an MLP layer for alignment, DINO surpasses CLIP in fine-grained related perception tasks. Building upon these observations, we propose a simple yet effective feature merging strategy, named COMM, that integrates CLIP and DINO with Multi-level features Merging, to enhance the visual capabilities of MLLMs. We evaluate COMM through comprehensive experiments on a wide range of benchmarks, including image captioning, visual question answering, visual grounding, and object hallucination. Experimental results demonstrate the superior performance of COMM compared to existing methods, showcasing its enhanced visual capabilities within MLLMs.

cs.CV↗

arXiv2023

scBiGNN: Bilevel Graph Representation Learning for Cell Type Classification from Single-cell RNA Sequencing Data

Single-cell RNA sequencing (scRNA-seq) technology provides high-throughput gene expression data to study the cellular heterogeneity and dynamics of complex organisms. Graph neural networks (GNNs) have been widely used for automatic cell type classification, which is a fundamental problem to solve in scRNA-seq analysis. However, existing methods do not sufficiently exploit both gene-gene and cell-cell relationships, and thus the true potential of GNNs is not realized. In this work, we propose a bilevel graph representation learning method, named scBiGNN, to simultaneously mine the relationships at both gene and cell levels for more accurate single-cell classification. Specifically, scBiGNN comprises two GNN modules to identify cell types. A gene-level GNN is established to adaptively learn gene-gene interactions and cell representations via the self-attention mechanism, and a cell-level GNN builds on the cell-cell graph that is constructed from the cell representations generated by the gene-level GNN. To tackle the scalability issue for processing a large number of cells, scBiGNN adopts an Expectation Maximization (EM) framework in which the two modules are alternately trained via the E-step and M-step to learn from each other. Through this interaction, the gene- and cell-level structural information is integrated to gradually enhance the classification performance of both GNN modules. Experiments on benchmark datasets demonstrate that our scBiGNN outperforms a variety of existing methods for cell type classification from scRNA-seq data.

cs.LG↗

arXiv2023

Spatial-Temporal DAG Convolutional Networks for End-to-End Joint Effective Connectivity Learning and Resting-State fMRI Classification

Building comprehensive brain connectomes has proved of fundamental importance in resting-state fMRI (rs-fMRI) analysis. Based on the foundation of brain network, spatial-temporal-based graph convolutional networks have dramatically improved the performance of deep learning methods in rs-fMRI time series classification. However, existing works either pre-define the brain network as the correlation matrix derived from the raw time series or jointly learn the connectome and model parameters without any topology constraint. These methods could suffer from degraded classification performance caused by the deviation from the intrinsic brain connectivity and lack biological interpretability of demonstrating the causal structure (i.e., effective connectivity) among brain regions. Moreover, most existing methods for effective connectivity learning are unaware of the downstream classification task and cannot sufficiently exploit useful rs-fMRI label information. To address these issues in an end-to-end manner, we model the brain network as a directed acyclic graph (DAG) to discover direct causal connections between brain regions and propose Spatial-Temporal DAG Convolutional Network (ST-DAGCN) to jointly infer effective connectivity and classify rs-fMRI time series by learning brain representations based on nonlinear structural equation model. The optimization problem is formulated into a continuous program and solved with score-based learning method via gradient descent. We evaluate ST-DAGCN on two public rs-fMRI databases. Experiments show that ST-DAGCN outperforms existing models by evident margins in rs-fMRI classification and simultaneously learns meaningful edges of effective connectivity that help understand brain activity patterns and pathological mechanisms in brain disease.

cs.LG↗

arXiv2023

AiluRus: A Scalable ViT Framework for Dense Prediction

Vision transformers (ViTs) have emerged as a prevalent architecture for vision tasks owing to their impressive performance. However, when it comes to handling long token sequences, especially in dense prediction tasks that require high-resolution input, the complexity of ViTs increases significantly. Notably, dense prediction tasks, such as semantic segmentation or object detection, emphasize more on the contours or shapes of objects, while the texture inside objects is less informative. Motivated by this observation, we propose to apply adaptive resolution for different regions in the image according to their importance. Specifically, at the intermediate layer of the ViT, we utilize a spatial-aware density-based clustering algorithm to select representative tokens from the token sequence. Once the representative tokens are determined, we proceed to merge other tokens into their closest representative token. Consequently, semantic similar tokens are merged together to form low-resolution regions, while semantic irrelevant tokens are preserved independently as high-resolution regions. This strategy effectively reduces the number of tokens, allowing subsequent layers to handle a reduced token sequence and achieve acceleration. We evaluate our proposed method on three different datasets and observe promising performance. For example, the "Segmenter ViT-L" model can be accelerated by 48% FPS without fine-tuning, while maintaining the performance. Additionally, our method can be applied to accelerate fine-tuning as well. Experimental results demonstrate that we can save 52% training time while accelerating 2.46 times FPS with only a 0.09% performance drop. The code is available at https://github.com/caddyless/ailurus/tree/main.

cs.CV↗

arXiv2023

MimicNorm: Weight Mean and Last BN Layer Mimic the Dynamic of Batch Normalization

Substantial experiments have validated the success of Batch Normalization (BN) Layer in benefiting convergence and generalization. However, BN requires extra memory and float-point calculation. Moreover, BN would be inaccurate on micro-batch, as it depends on batch statistics. In this paper, we address these problems by simplifying BN regularization while keeping two fundamental impacts of BN layers, i.e., data decorrelation and adaptive learning rate. We propose a novel normalization method, named MimicNorm, to improve the convergence and efficiency in network training. MimicNorm consists of only two light operations, including modified weight mean operations (subtract mean values from weight parameter tensor) and one BN layer before loss function (last BN layer). We leverage the neural tangent kernel (NTK) theory to prove that our weight mean operation whitens activations and transits network into the chaotic regime like BN layer, and consequently, leads to an enhanced convergence. The last BN layer provides autotuned learning rates and also improves accuracy. Experimental results show that MimicNorm achieves similar accuracy for various network structures, including ResNets and lightweight networks like ShuffleNet, with a reduction of about 20% memory consumption. The code is publicly available at https://github.com/Kid-key/MimicNorm.

cs.LG↗

arXiv2023

Prune Spatio-temporal Tokens by Semantic-aware Temporal Accumulation

Transformers have become the primary backbone of the computer vision community due to their impressive performance. However, the unfriendly computation cost impedes their potential in the video recognition domain. To optimize the speed-accuracy trade-off, we propose Semantic-aware Temporal Accumulation score (STA) to prune spatio-temporal tokens integrally. STA score considers two critical factors: temporal redundancy and semantic importance. The former depicts a specific region based on whether it is a new occurrence or a seen entity by aggregating token-to-token similarity in consecutive frames while the latter evaluates each token based on its contribution to the overall prediction. As a result, tokens with higher scores of STA carry more temporal redundancy as well as lower semantics thus being pruned. Based on the STA score, we are able to progressively prune the tokens without introducing any additional parameters or requiring further re-training. We directly apply the STA module to off-the-shelf ViT and VideoSwin backbones, and the empirical results on Kinetics-400 and Something-Something V2 achieve over 30% computation reduction with a negligible ~0.2% accuracy drop. The code is released at https://github.com/Mark12Ding/STA.

cs.CV↗

arXiv2023

ActionPrompt: Action-Guided 3D Human Pose Estimation With Text and Pose Prompting

Recent 2D-to-3D human pose estimation (HPE) utilizes temporal consistency across sequences to alleviate the depth ambiguity problem but ignore the action related prior knowledge hidden in the pose sequence. In this paper, we propose a plug-and-play module named Action Prompt Module (APM) that effectively mines different kinds of action clues for 3D HPE. The highlight is that, the mining scheme of APM can be widely adapted to different frameworks and bring consistent benefits. Specifically, we first present a novel Action-related Text Prompt module (ATP) that directly embeds action labels and transfers the rich language information in the label to the pose sequence. Besides, we further introduce Action-specific Pose Prompt module (APP) to mine the position-aware pose pattern of each action, and exploit the correlation between the mined patterns and input pose sequence for further pose refinement. Experiments show that APM can improve the performance of most video-based 2D-to-3D HPE frameworks by a large margin.

cs.CV↗