Binarization demonstrates significant advantages in resource-limited devices, particularly for recent Vision Transformers (ViTs). To alleviate performance degradation during binarization, Knowledge Distillation (KD) is a crucial and beneficial technique. However, we identify that distilling ViTs in conventional methods overlooks high-frequency information, which loses fine-grained features and results in performance degradation. To address this challenge, we introduce a plug-and-play Binarized Frequency-enhanced Distillation (BFD) to preserve high-frequency information effectively. Specifically, High-Frequency Enhanced Distillation (HFED) is proposed to transmit attention maps from teacher to student in frequency domain, which enhances high-frequency information. Additionally, based on the finding that the frequency proportion varies across different layers, we further propose Progressive Frequency Partitioning (PFP) to distinguish frequency flexibly. Extensive experiments demonstrate the effectiveness of our BFD across a set of ViT variants. We demonstrate superior performance of 51.65% over ViT-B on TinyImageNet, and outperform SOTA by a substantial margin of 20.70% on ImageNet.

Binarized Vision Transformers (BiViTs) aim to facilitate the efficient and lightweight utilization of Vision Transformers (ViTs) on devices with limited computational resources. Yet, the current approach to binarizing ViT leads to a substantial performance decrease compared to the full-precision model, posing obstacles to practical deployment. By empirical study, we reveal that spatial interaction (SI) is a critical factor that impacts performance due to lack of token-level correlation, but previous work ignores this factor. To this end, we design a ViT binarization approach dubbed SI-BiViT to incorporate spatial interaction in the binarization process. Specifically, an SI module is placed alongside the Multi-Layer Perceptron (MLP) module to formulate the dual-branch structure. This structure not only leverages knowledge from pre-trained ViTs by distilling over the original MLP, but also enhances spatial interaction via the introduced SI module. Correspondingly, we design a decoupled training strategy to train these two branches more effectively. Importantly, our SI-BiViT is orthogonal to existing Binarized ViTs approaches and can be directly plugged. Extensive experiments demonstrate the strong flexibility and effectiveness of SI-BiViT by plugging our method into four classic ViT backbones in supporting three downstream tasks, including classification, detection, and segmentation. In particular, SI-BiViT enhances the classification performance of binarized ViTs by an average of 10.52% in Top-1 accuracy compared to the previous state-of-the-art. Codes are available at https://github.com/VL-Group/SI-BiViT

The focus of this study is on Unsupervised Continual Learning (UCL), as it presents an alternative to Supervised Continual Learning which needs high-quality manual labeled data. The experiments under UCL paradigm indicate a phenomenon where the results on the first few tasks are suboptimal. This phenomenon can render the model inappropriate for practical applications. To address this issue, after analyzing the phenomenon and identifying the lack of diversity as a vital factor, we propose a method named Codebook for Unsupervised Continual Learning (CUCL) which promotes the model to learn discriminative features to complete the class boundary. Specifically, we first introduce a Product Quantization to inject diversity into the representation and apply a cross quantized contrastive loss between the original representation and the quantized one to capture discriminative information. Then, based on the quantizer, we propose a effective Codebook Rehearsal to address catastrophic forgetting. This study involves conducting extensive experiments on CIFAR100, TinyImageNet, and MiniImageNet benchmark datasets. Our method significantly boosts the performances of supervised and unsupervised methods. For instance, on TinyImageNet, our method led to a relative improvement of 12.76% and 7% when compared with Simsiam and BYOL, respectively. Codes are publicly available at https://github.com/zackschen/CUCL

Cross-modal Retrieval methods build similarity relations between vision and language modalities by jointly learning a common representation space. However, the predictions are often unreliable due to the Aleatoric uncertainty, which is induced by low-quality data, e.g., corrupt images, fast-paced videos, and non-detailed texts. In this paper, we propose a novel Prototype-based Aleatoric Uncertainty Quantification (PAU) framework to provide trustworthy predictions by quantifying the uncertainty arisen from the inherent data ambiguity. Concretely, we first construct a set of various learnable prototypes for each modality to represent the entire semantics subspace. Then Dempster-Shafer Theory and Subjective Logic Theory are utilized to build an evidential theoretical framework by associating evidence with Dirichlet Distribution parameters. The PAU model induces accurate uncertainty and reliable predictions for cross-modal retrieval. Extensive experiments are performed on four major benchmark datasets of MSR-VTT, MSVD, DiDeMo, and MS-COCO, demonstrating the effectiveness of our method. The code is accessible at https://github.com/leolee99/PAU.

Lately, video-language pre-training and text-video retrieval have attracted significant attention with the explosion of multimedia data on the Internet. However, existing approaches for video-language pre-training typically limit the exploitation of the hierarchical semantic information in videos, such as frame semantic information and global video semantic information. In this work, we present an end-to-end pre-training network with Hierarchical Matching and Momentum Contrast named HMMC. The key idea is to explore the hierarchical semantic information in videos via multilevel semantic matching between videos and texts. This design is motivated by the observation that if a video semantically matches a text (can be a title, tag or caption), the frames in this video usually have semantic connections with the text and show higher similarity than frames in other videos. Hierarchical matching is mainly realized by two proxy tasks: Video-Text Matching (VTM) and Frame-Text Matching (FTM). Another proxy task: Frame Adjacency Matching (FAM) is proposed to enhance the single visual modality representations while training from scratch. Furthermore, momentum contrast framework was introduced into HMMC to form a multimodal momentum contrast framework, enabling HMMC to incorporate more negative samples for contrastive learning which contributes to the generalization of representations. We also collected a large-scale Chinese video-language dataset (over 763k unique videos) named CHVTT to explore the multilevel semantic connections between videos and texts. Experimental results on two major Text-video retrieval benchmark datasets demonstrate the advantages of our methods. We release our code at https://github.com/cheetah003/HMMC.

Existing supervised quantization methods usually learn the quantizers from pair-wise, triplet, or anchor-based losses, which only capture their relationship locally without aligning them globally. This may cause an inadequate use of the entire space and a severe intersection among different semantics, leading to inferior retrieval performance. Furthermore, to enable quantizers to learn in an end-to-end way, current practices usually relax the non-differentiable quantization operation by substituting it with softmax, which unfortunately is biased, leading to an unsatisfying suboptimal solution. To address the above issues, we present Spherical Centralized Quantization (SCQ), which contains a Priori Knowledge based Feature (PKFA) module for the global alignment of feature vectors, and an Annealing Regulation Semantic Quantization (ARSQ) module for low-biased optimization. Specifically, the PKFA module first applies Semantic Center Allocation (SCA) to obtain semantic centers based on prior knowledge, and then adopts Centralized Feature Alignment (CFA) to gather feature vectors based on corresponding semantic centers. The SCA and CFA globally optimize the inter-class separability and intra-class compactness, respectively. After that, the ARSQ module performs a partial-soft relaxation to tackle biases, and an Annealing Regulation Quantization loss for further addressing the local optimal solution. Experimental results show that our SCQ outperforms state-of-the-art algorithms by a large margin (2.1%, 3.6%, 5.5% mAP respectively) on CIFAR-10, NUS-WIDE, and ImageNet with a code length of 8 bits. Codes are publicly available:https://github.com/zzb111/Spherical-Centralized-Quantization.

Multi-Codebook Quantization (MCQ) is a generalized version of existing codebook-based quantizations for Approximate Nearest Neighbor (ANN) search. Specifically, MCQ picks one codeword for each sub-codebook independently and takes the sum of picked codewords to approximate the original vector. The objective function involves no constraints, therefore, MCQ theoretically has the potential to achieve the best performance because solutions of other codebook-based quantization methods are all covered by MCQ's solution space under the same codebook size setting. However, finding the optimal solution to MCQ is proved to be NP-hard due to its encoding process, i.e., converting an input vector to a binary code. To tackle this, researchers apply constraints to it to find near-optimal solutions or employ heuristic algorithms that are still time-consuming for encoding. Different from previous approaches, this paper takes the first attempt to find a deep solution to MCQ. The encoding network is designed to be as simple as possible, so the very complex encoding problem becomes simply a feed-forward. Compared with other methods on three datasets, our method shows state-of-the-art performance. Notably, our method is 11× - 38× faster than heuristic algorithms for encoding, which makes it more practical for the real scenery of large-scale retrieval. Our code is publicly available: https://github.com/DeepMCQ/DeepQ.

Modeling latent variables with priors and hyperpriors is an essential problem in variational image compression. Formally, trade-off between rate and distortion is handled well if priors and hyperpriors precisely describe latent variables. Current practices only adopt univariate priors and process each variable individually. However, we find inter-correlations and intra-correlations exist when observing latent variables in a vectorized perspective. These findings reveal visual redundancies to improve rate-distortion performance and parallel processing ability to speed up compression. This encourages us to propose a novel vectorized prior. Specifically, a multivariate Gaussian mixture is proposed with means and covariances to be estimated. Then, a novel probabilistic vector quantization is utilized to effectively approximate means, and remaining covariances are further induced to a unified mixture and solved by cascaded estimation without context models involved. Furthermore, code books involved in quantization are extended to multi-codebooks for complexity reduction, which formulates an efficient compression procedure. Extensive experiments on benchmark datasets against state-of-the-art indicate our model has better rate-distortion performance and an impressive 3.18x compression speed up, giving us the ability to perform real-time, high-quality variational image compression in practice. Our source code is publicly available at https://github.com/xiaosu-zhu/McQuic.

As a crucial approach for compact representation learning, hashing has achieved great success in effectiveness and efficiency. Numerous heuristic Hamming space metric learning objectives are designed to obtain high-quality hash codes. Nevertheless, a theoretical analysis of criteria for learning good hash codes remains largely unexploited. In this paper, we prove that inter-class distinctiveness and intra-class compactness among hash codes determine the lower bound of hash codes' performance. Promoting these two characteristics could lift the bound and improve hash learning. We then propose a surrogate model to fully exploit the above objective by estimating the posterior of hash codes and controlling it, which results in a low-bias optimization. Extensive experiments reveal the effectiveness of the proposed method. By testing on a series of hash-models, we obtain performance improvements among all of them, with an up to $26.5\%$ increase in mean Average Precision and an up to $20.5\%$ increase in accuracy. Our code is publicly available at https://github.com/VL-Group/LBHash.

Despite the success of single-domain person re-identification (ReID), current supervised models degrade dramatically when deployed to unseen domains, mainly due to the discrepancy across cameras. To tackle this issue, we propose an Adversarial Disentangling Learning (ADL) framework to decouple camera-related and ID-related features, which can be readily used for camera-agnostic person ReID. ADL adopts a discriminative way instead of the mainstream generative styles in disentangling methods, eg., GAN or VAE based, because for person ReID task only the information to discriminate IDs is needed, and more information to generate images are redundant and may be noisy. Specifically, our model involves a feature separation module that encodes images into two separate feature spaces and a disentangled feature learning module that performs adversarial training to minimize mutual information. We design an effective solution to approximate and minimize mutual information by transforming it into a discrimination problem. The two modules are co-designed to obtain strong generalization ability by only using source dataset. Extensive experiments on three public benchmarks show that our method outperforms the state-of-the-art generalizable person ReID model by a large margin. Our code is publicly available at https://github.com/luckyaci/ADL_ReID.

Unsupervised video quantization is to compress the original videos to compact binary codes so that video retrieval can be conducted in an efficient way. In this paper, we make a first attempt to combine quantization method with video retrieval called 3D-UVQ, which obtains high retrieval accuracy with low storage cost. In the proposed framework, we address two main problems: 1) how to design an effective pipeline to perceive video contextual information for video features extraction; and 2) how to quantize these features for efficient retrieval. To tackle these problems, we propose a 3D self-attention module to exploit the spatial and temporal contextual information, where each pixel is influenced by its surrounding pixels. By taking a further recurrent operation, each pixel can finally capture the global context from all pixels. Then, we propose gradient-based residual quantization which consists of several quantization blocks to approximate the features gradually. Extensive experimental results on three benchmark datasets demonstrate that our method significantly outperforms the state-of-the-arts. Ablation study shows that both the 3D self-attention module and the gradient-based residual quantization can improve the performance of retrieval. Our model is publicly available at https://github.com/brownwolf/3D-UVQ.

Product Quantization (PQ) has long been a mainstream for generating an exponentially large codebook at very low memory/time cost. Despite its success, PQ is still tricky for the decomposition of high-dimensional vector space, and the retraining of model is usually unavoidable when the code length changes. In this work, we propose a deep progressive quantization (DPQ) model, as an alternative to PQ, for large scale image retrieval. DPQ learns the quantization codes sequentially and approximates the original feature space progressively. Therefore, we can train the quantization codes with different code lengths simultaneously. Specifically, we first utilize the label information for guiding the learning of visual features, and then apply several quantization blocks to progressively approach the visual features. Each quantization block is designed to be a layer of a convolutional neural network, and the whole framework can be trained in an end-to-end manner. Experimental results on the benchmark datasets show that our model significantly outperforms the state-of-the-art for image retrieval. Our model is trained once for different code lengths and therefore requires less computation time. Additional ablation study demonstrates the effect of each component of our proposed model. Our code is released at https://github.com/cfm-uestc/DPQ.

Quantization has been an effective technology in ANN (approximate nearest neighbour) search due to its high accuracy and fast search speed. To meet the requirement of different applications, there is always a trade-off between retrieval accuracy and speed, reflected by variable code lengths. However, to encode the dataset into different code lengths, existing methods need to train several models, where each model can only produce a specific code length. This incurs a considerable training time cost, and largely reduces the flexibility of quantization methods to be deployed in real applications. To address this issue, we propose a Deep Recurrent Quantization (DRQ) architecture which can generate sequential binary codes. To the end, when the model is trained, a sequence of binary codes can be generated and the code length can be easily controlled by adjusting the number of recurrent iterations. A shared codebook and a scalar factor is designed to be the learnable weights in the deep recurrent quantization block, and the whole framework can be trained in an end-to-end manner. As far as we know, this is the first quantization method that can be trained once and generate sequential binary codes. Experimental results on the benchmark datasets show that our model achieves comparable or even better performance compared with the state-of-the-art for image retrieval. But it requires significantly less number of parameters and training times. Our code is published online: https://github.com/cfm-uestc/DRQ.