Masked image modeling (MIM) pre-training for large-scale vision transformers (ViTs) in computer vision has enabled promising downstream performance on top of the learned self-supervised ViT features. In this paper, we question if the extremely simple ViTs' fine-tuning performance with a small-scale architecture can also benefit from this pre-training paradigm, which is considerably less studied yet in contrast to the well-established lightweight architecture design methodology with sophisticated components introduced. By carefully adapting various typical MIM pre-training methods to this lightweight regime and comparing them with the contrastive learning (CL) pre-training on various downstream image classification and dense prediction tasks, we systematically observe different behaviors between MIM and CL with respect to the downstream fine-tuning data scales. Furthermore, we analyze the frozen features under linear probing evaluation and also the layer representation similarities and attention maps across the obtained models, which clearly show the inferior learning of MIM pre-training on higher layers, leading to unsatisfactory fine-tuning performance on data-insufficient downstream tasks. This finding is naturally a guide to choosing appropriate distillation strategies during pre-training to solve the above deterioration problem. Extensive experiments on various vision tasks demonstrate the effectiveness of our observation-analysis-solution flow. In particular, our pre-training with distillation on pure lightweight ViTs with vanilla/hierarchical design (5.7M/6.5M) can achieve 79.4%/78.9% top-1 accuracy on ImageNet-1K. It also enables SOTA performance on the ADE20K semantic segmentation task (42.8% mIoU) and LaSOT visual tracking task (66.1% AUC) in the lightweight regime. The latter even surpasses all the current SOTA lightweight CPU-realtime trackers.
Recent progress in pre-trained diffusion models and 3D generation have spurred interest in 4D content creation. However, achieving high-fidelity 4D generation with spatial-temporal consistency remains a challenge. In this work, we propose STAG4D, a novel framework that combines pre-trained diffusion models with dynamic 3D Gaussian splatting for high-fidelity 4D generation. Drawing inspiration from 3D generation techniques, we utilize a multi-view diffusion model to initialize multi-view images anchoring on the input video frames, where the video can be either real-world captured or generated by a video diffusion model. To ensure the temporal consistency of the multi-view sequence initialization, we introduce a simple yet effective fusion strategy to leverage the first frame as a temporal anchor in the self-attention computation. With the almost consistent multi-view sequences, we then apply the score distillation sampling to optimize the 4D Gaussian point cloud. The 4D Gaussian spatting is specially crafted for the generation task, where an adaptive densification strategy is proposed to mitigate the unstable Gaussian gradient for robust optimization. Notably, the proposed pipeline does not require any pre-training or fine-tuning of diffusion networks, offering a more accessible and practical solution for the 4D generation task. Extensive experiments demonstrate that our method outperforms prior 4D generation works in rendering quality, spatial-temporal consistency, and generation robustness, setting a new state-of-the-art for 4D generation from diverse inputs, including text, image, and video.
In this paper, we propose a new image-based visual place recognition (VPR) framework by exploiting the structural cues in bird's-eye view (BEV) from a single monocular camera. The motivation arises from two key observations about VPR: 1) For the methods based on both camera and LiDAR sensors, the integration of LiDAR in robotic systems has led to increased expenses, while the alignment of data between different sensors is also a major challenge. 2) Other image-/camera-based methods, involving integrating RGB images and their derived variants (e.g., pseudo depth images, pseudo 3D point clouds), exhibit several limitations, such as the failure to effectively exploit the explicit spatial relationships between different objects. To tackle the above issues, we design a new BEV-enhanced VPR framework, nemely BEV2PR, which can generate a composite descriptor with both visual cues and spatial awareness solely based on a single camera. For the visual cues, any popular aggregation module for RGB global features can be integrated into our framework. The key points lie in: 1) We use BEV segmentation features as an explicit source of structural knowledge in constructing global features. 2) The lower layers of the pre-trained backbone from BEV map generation are shared for visual and structural streams in VPR, facilitating the learning of fine-grained local features in the visual stream. 3) The complementary visual features and structural features can jointly enhance VPR performance. Our BEV2PR framework enables consistent performance improvements over several popular camera-based VPR aggregation modules when integrating them. The experiments on our collected VPR-NuScenes dataset demonstrate an absolute gain of 2.47% on Recall@1 for the strong Conv-AP baseline to achieve the best performance in our setting, and notably, a 18.06% gain on the hard set.
Due to the diversity of assessment requirements in various application scenarios for the IQA task, existing IQA methods struggle to directly adapt to these varied requirements after training. Thus, when facing new requirements, a typical approach is fine-tuning these models on datasets specifically created for those requirements. However, it is time-consuming to establish IQA datasets. In this work, we propose a Prompt-based IQA (PromptIQA) that can directly adapt to new requirements without fine-tuning after training. On one hand, it utilizes a short sequence of Image-Score Pairs (ISP) as prompts for targeted predictions, which significantly reduces the dependency on the data requirements. On the other hand, PromptIQA is trained on a mixed dataset with two proposed data augmentation strategies to learn diverse requirements, thus enabling it to effectively adapt to new requirements. Experiments indicate that the PromptIQA outperforms SOTA methods with higher performance and better generalization. The code will be available.
In vision-language pre-training (VLP), masked image modeling (MIM) has recently been introduced for fine-grained cross-modal alignment. However, in most existing methods, the reconstruction targets for MIM lack high-level semantics, and text is not sufficiently involved in masked modeling. These two drawbacks limit the effect of MIM in facilitating cross-modal semantic alignment. In this work, we propose a semantics-enhanced cross-modal MIM framework (SemMIM) for vision-language representation learning. Specifically, to provide more semantically meaningful supervision for MIM, we propose a local semantics enhancing approach, which harvest high-level semantics from global image features via self-supervised agreement learning and transfer them to local patch encodings by sharing the encoding space. Moreover, to achieve deep involvement of text during the entire MIM process, we propose a text-guided masking strategy and devise an efficient way of injecting textual information in both masked modeling and reconstruction target acquisition. Experimental results validate that our method improves the effectiveness of the MIM task in facilitating cross-modal semantic alignment. Compared to previous VLP models with similar model size and data scale, our SemMIM model achieves state-of-the-art or competitive performance on multiple downstream vision-language tasks.
In video-text retrieval, most existing methods adopt the dual-encoder architecture for fast retrieval, which employs two individual encoders to extract global latent representations for videos and texts. However, they face challenges in capturing fine-grained semantic concepts. In this work, we propose the UNIFY framework, which learns lexicon representations to capture fine-grained semantics and combines the strengths of latent and lexicon representations for video-text retrieval. Specifically, we map videos and texts into a pre-defined lexicon space, where each dimension corresponds to a semantic concept. A two-stage semantics grounding approach is proposed to activate semantically relevant dimensions and suppress irrelevant dimensions. The learned lexicon representations can thus reflect fine-grained semantics of videos and texts. Furthermore, to leverage the complementarity between latent and lexicon representations, we propose a unified learning scheme to facilitate mutual learning via structure sharing and self-distillation. Experimental results show our UNIFY framework largely outperforms previous video-text retrieval methods, with 4.8% and 8.2% Recall@1 improvement on MSR-VTT and DiDeMo respectively.
Due to the subjective nature of image quality assessment (IQA), assessing which image has better quality among a sequence of images is more reliable than assigning an absolute mean opinion score for an image. Thus, IQA models are evaluated by global correlation consistency (GCC) metrics like PLCC and SROCC, rather than mean opinion consistency (MOC) metrics like MAE and MSE. However, most existing methods adopt MOC metrics to define their loss functions, due to the infeasible computation of GCC metrics during training. In this work, we construct a novel loss function and network to exploit Global-correlation and Mean-opinion Consistency, forming a GMC-IQA framework. Specifically, we propose a novel GCC loss by defining a pairwise preference-based rank estimation to solve the non-differentiable problem of SROCC and introducing a queue mechanism to reserve previous data to approximate the global results of the whole data. Moreover, we propose a mean-opinion network, which integrates diverse opinion features to alleviate the randomness of weight learning and enhance the model robustness. Experiments indicate that our method outperforms SOTA methods on multiple authentic datasets with higher accuracy and generalization. We also adapt the proposed loss to various networks, which brings better performance and more stable training.
In the rapidly evolving domain of digital content generation, the focus has shifted from text-to-image (T2I) models to more advanced video diffusion models, notably text-to-video (T2V) and image-to-video (I2V). This paper addresses the intricate challenge posed by I2V: converting static images into dynamic, lifelike video sequences while preserving the original image fidelity. Traditional methods typically involve integrating entire images into diffusion processes or using pretrained encoders for cross attention. However, these approaches often necessitate altering the fundamental weights of T2I models, thereby restricting their reusability. We introduce a novel solution, namely I2V-Adapter, designed to overcome such limitations. Our approach preserves the structural integrity of T2I models and their inherent motion modules. The I2V-Adapter operates by processing noised video frames in parallel with the input image, utilizing a lightweight adapter module. This module acts as a bridge, efficiently linking the input to the model's self-attention mechanism, thus maintaining spatial details without requiring structural changes to the T2I model. Moreover, I2V-Adapter requires only a fraction of the parameters of conventional models and ensures compatibility with existing community-driven T2I models and controlling tools. Our experimental results demonstrate I2V-Adapter's capability to produce high-quality video outputs. This performance, coupled with its versatility and reduced need for trainable parameters, represents a substantial advancement in the field of AI-driven video generation, particularly for creative applications.
Diverse video captioning aims to generate a set of sentences to describe the given video in various aspects. Mainstream methods are trained with independent pairs of a video and a caption from its ground-truth set without exploiting the intra-set relationship, resulting in low diversity of generated captions. Different from them, we formulate diverse captioning into a semantic-concept-guided set prediction (SCG-SP) problem by fitting the predicted caption set to the ground-truth set, where the set-level relationship is fully captured. Specifically, our set prediction consists of two synergistic tasks, i.e., caption generation and an auxiliary task of concept combination prediction providing extra semantic supervision. Each caption in the set is attached to a concept combination indicating the primary semantic content of the caption and facilitating element alignment in set prediction. Furthermore, we apply a diversity regularization term on concepts to encourage the model to generate semantically diverse captions with various concept combinations. These two tasks share multiple semantics-specific encodings as input, which are obtained by iterative interaction between visual features and conceptual queries. The correspondence between the generated captions and specific concept combinations further guarantees the interpretability of our model. Extensive experiments on benchmark datasets show that the proposed SCG-SP achieves state-of-the-art (SOTA) performance under both relevance and diversity metrics.
Recently, the transformer has enabled the speed-oriented trackers to approach state-of-the-art (SOTA) performance with high-speed thanks to the smaller input size or the lighter feature extraction backbone, though they still substantially lag behind their corresponding performance-oriented versions. In this paper, we demonstrate that it is possible to narrow or even close this gap while achieving high tracking speed based on the smaller input size. To this end, we non-uniformly resize the cropped image to have a smaller input size while the resolution of the area where the target is more likely to appear is higher and vice versa. This enables us to solve the dilemma of attending to a larger visual field while retaining more raw information for the target despite a smaller input size. Our formulation for the non-uniform resizing can be efficiently solved through quadratic programming (QP) and naturally integrated into most of the crop-based local trackers. Comprehensive experiments on five challenging datasets based on two kinds of transformer trackers, \ie, OSTrack and TransT, demonstrate consistent improvements over them. In particular, applying our method to the speed-oriented version of OSTrack even outperforms its performance-oriented counterpart by 0.6% AUC on TNL2K, while running 50% faster and saving over 55% MACs. Codes and models are available at https://github.com/Kou-99/ZoomTrack.