Pre-trained vision transformers (ViTs) have demonstrated remarkable capability in learning semantically rich image representations. However, their underlying plain architectures yield low-resolution feature maps, lacking essential fine-grained spatial details required for dense prediction tasks. To better transfer the learned visual features, we present ViT-Hook, a novel hybrid backbone compatible with plain ViTs that effectively bridges the gap between global semantic understanding and local spatial encodings. Specifically, our method aims to broaden the scope and impact of ViT from the following perspectives: (1) We propose a simple transformer-decoder-inspired hook module that receives hierarchical CNN features as spatial queries and interacts with expressive ViT features from large-scale pre-training, therefore instantiating general-purpose representations into task-suited ones. (2) ViT-Hook is a plug-and-play solution for powerful vision foundation models, such as DINOv2 and RADIO. In this case, we find that only partially fine-tuning several intermediate ViT layers can outperform previous full fine-tuning methods, while substantially reducing compute and memory burdens with most parameters frozen. (3) We evaluate ViT-Hook with various pre-trained sources on multiple dense prediction tasks, including semantic segmentation, instance segmentation, and object detection. Notably, tested on the unified UperNet and Mask R-CNN frameworks, our ViT-Hook surpasses state-of-the-art by a large margin, achieving 59.7 (+4.7) mIoU on ADE20K val, 55.0 (+3.6) box AP and 48.5 (+3.3) mask AP on COCO val2017. • We propose ViT-Hook, a hybrid backbone that effectively enhances Vision Transformer performance on various dense prediction tasks. • The proposed spatial query and hook modules are lightweight yet powerful, achieving competitive results compared to SoTA on widely used benchmarks. • We introduce a novel partial fine-tuning strategy, which outperforms full fine-tuning while using only a fraction of compute and memory. • We validate the generalizability of ViT-Hook on multiple types of upstream pre-training methods, including the most recent vision foundation models.

Achieving the serializable isolation level is costly. Recent studies have revealed that adjusting specific query patterns within the workload can still achieve serializability, even at lower isolation levels. Nevertheless, these studies typically overlook the trade-off between the performance advantages of lower isolation levels and the overhead required to maintain serializability, potentially leading to suboptimal isolation level choices that fail to maximize performance. In this paper, we present TxnSails, a middle-tier solution designed to achieve serializable scheduling with self-adaptive isolation level selection. First, TxnSails incorporates a unified concurrency control algorithm that achieves serializability at lower isolation levels with minimal overhead. Second, TxnSails employs a deep learning method to characterize the trade-off between the performance benefits and overhead associated with lower isolation levels, thus predicting the optimal isolation level. Finally, TxnSails implements a cross-isolation validation mechanism to ensure serializability during real-time isolation level transitions. Extensive experiments demonstrate that TxnSails outperforms existing solutions by up to 26.7× and PostgreSQL 's serializable isolation level by up to 4.8×.

Visual appearance is considered to be the most important cue to understand images for cross-modal retrieval, while sometimes the scene text appearing in images can provide valuable information to understand the visual semantics. Most of existing cross-modal retrieval approaches ignore the usage of scene text information and directly adding this information may lead to performance degradation in scene text free scenarios. To address this issue, we propose a full transformer architecture to unify these cross-modal retrieval scenarios in a single Vision and Scene Text Aggregation framework (ViSTA). Specifically, ViSTA utilizes transformer blocks to directly encode image patches and fuse scene text embedding to learn an aggregated visual representation for cross-modal retrieval. To tackle the modality missing problem of scene text, we propose a novel fusion token based transformer aggregation approach to exchange the necessary scene text information only through the fusion token and concentrate on the most important features in each modality. To further strengthen the visual modality, we develop dual contrastive learning losses to embed both image-text pairs and fusion-text pairs into a common cross-modal space. Compared to existing methods, ViSTA enables to aggregate relevant scene text semantics with visual appearance, and hence improve results under both scene text free and scene text aware scenarios. Experimental results show that ViSTA outperforms other methods by at least 8.4% at Recall@ 1 for scene text aware retrieval task. Compared with state-of-the-art scene text free retrieval methods, ViSTA can achieve better accuracy on Flicker30K and MSCOCO while running at least three times faster during the inference stage, which validates the effectiveness of the proposed framework.

Most existing text reading benchmarks make it difficult to evaluate the performance of more advanced deep learning models in large vocabularies due to the limited amount of training data. To address this issue, we introduce a new large-scale text reading benchmark dataset named Chinese Street View Text (C-SVT) with 430,000 street view images, which is at least 14 times as large as the existing Chinese text reading benchmarks. To recognize Chinese text in the wild while keeping large-scale datasets labeling cost-effective, we propose to annotate one part of the C-SVT dataset (30,000 images) in locations and text labels as full annotations and add 400,000 more images, where only the corresponding text-of-interest in the regions is given as weak annotations. To exploit the rich information from the weakly annotated data, we design a text reading network in a partially supervised learning framework, which enables to localize and recognize text, learn from fully and weakly annotated data simultaneously. To localize the best matched text proposals from weakly labeled images, we propose an online proposal matching module incorporated in the whole model, spotting the keyword regions by sharing parameters for end-to-end training. Compared with fully supervised training algorithms, this model can improve the end-to-end recognition performance remarkably by 4.03% in F-score at the same labeling cost. The proposed model can also achieve state-of-the-art results on the ICDAR 2017-RCTW dataset, which demonstrates the effectiveness of the proposed partially supervised learning framework.

This paper reports the ICDAR2019 Robust Reading Challenge on Arbitrary-Shaped Text - RRC-ArT that consists of three major challenges: i) scene text detection, ii) scene text recognition, and iii) scene text spotting. A total of 78 submissions from 46 unique teams/individuals were received for this competition. The top performing score of each challenge is as follows: i) T1 - 82.65%, ii) T2.1 - 74.3%, iii) T2.2 - 85.32%, iv) T3.1 - 53.86%, and v) T3.2 - 54.91%. Apart from the results, this paper also details the ArT dataset, tasks description, evaluation metrics and participants' methods. The dataset, the evaluation kit as well as the results are publicly available at the challenge website.

Robust text reading from street view images provides valuable information for various applications. Performance improvement of existing methods in such a challenging scenario heavily relies on the amount of fully annotated training data, which is costly and in-efficient to obtain. To scale up the amount of training data while keeping the labeling procedure cost-effective, this competition introduces a new challenge on Large-scale Street View Text with Partial Labeling (LSVT), providing 5,0000 and 400,000 images in full and weak annotations, respectively. This competition aims to explore the abilities of state-of-the-art methods to detect and recognize text instances from large-scale street view images, closing gaps between research benchmarks and real applications. During the competition period, a total number of 41 teams participate in the two tasks with 132 valid submissions, i.e., text detection and end-to-end text spotting. This paper includes dataset descriptions, task definitions, evaluation protocols and results summaries of ICDAR 2019-LSVT challenge.

Motivated by the attention on salient objects, conventional region-of-interest (ROI)-based image coding approaches attempt to assign more bits to ROIs and fewer bits to other regions. Thus, the perceptual quality of salient object regions is improved by sacrificing the quality of non-ROI regions with unpleasant artifacts. To address this issue, we concentrate on the efficient compression of object-centered images by encoding salient objects and background features separately. To fully recover the object and background, we propose a hierarchical exemplar-based matching-synthesis (HEMS) approach to reconstruct the image from exemplars. In the proposed framework, once the salient object regions are encoded, only the quantized color features and local descriptors of the background are kept, achieving bit-rate reduction. To make it possible and practical to reconstruct background regions, the hierarchical framework is designed in three layers, including relevant image search, patch candidates matching, and distortion optimized image synthesis. In the hierarchical framework, firstly, image search from an external database returns relevant images, limiting the search space to a feasible number of patch candidates. Secondly, patches are matched by color features to select the appropriate candidates. Finally, the distortion optimized image synthesis further makes it possible to automatically choose the most suitable texture sample, and seamlessly reconstruct the image. Compared to the conventional ROI-based image coding schemes, the proposed approach can achieve better visual quality on both ROI and background regions.

Principal component analysis (PCA) is widely used to extract features and reduce dimensionality in various computer vision and image/video processing tasks. Conventional approaches either lack robustness to outliers and corrupted data or are designed for one-dimensional signals. To address this problem, we propose a robust PCA model for two-dimensional images incorporating structured sparse priors, referred to as structured sparse 2D-PCA. This robust model considers the prior of structured and grouped pixel values in two dimensions. As the proposed formulation is jointly nonconvex and nonsmooth, which is difficult to tackle by joint optimization, we develop a two-stage alternating minimization approach to solve the problem. This approach iteratively learns the projection matrices by bidirectional decomposition and utilizes the proximal method to obtain the structured sparse outliers. By considering the structured sparsity prior, the proposed model becomes less sensitive to noisy data and outliers in two dimensions. Moreover, the computational cost indicates that the robust two-dimensional model is capable of processing quarter common intermediate format video in real time, as well as handling large-size images and videos, which is often intractable with other robust PCA approaches that involve image-to-vector conversion. Experimental results on robust face reconstruction, video background subtraction data set, and real-world videos show the effectiveness of the proposed model compared with conventional 2D-PCA and other robust PCA algorithms.

In this brief we propose a multisample sparse representation (MSR)-based online dictionary-learning approach to encode images more efficiently. To minimize the reconstructed error while handling a variety of image samples, we develop a multisample sparse representation method capable of obtaining sparser coefficients combined with learning dictionaries on-the-fly. With a well-learned dictionary, we further derive an MSR-based image coding approach to encode the quantized sparse coefficients with reduced reconstructed errors. Experimental results demonstrate rapid convergence of the proposed dictionary-learning algorithm and improved rate-distortion performance over other competitive image compression schemes both subjectively and quantitatively, validating the effectiveness of the proposed approach.

To extract two-dimensional principal components from image samples while being insensitive to outliers, we propose a robust model for two-dimensional principal component analysis (robust 2D-PCA) by regularizing sparse penalty term. Moveover, we develop a novel iterative algorithm for robust 2D-PCA via alternating optimization, learning the projection matrices by bi-directional decomposition. To further speed up the iteration, we develop an alternating greedy approach, minimizing over the low-dimensional feature matrix and the sparse error matrix. Experimental results on dynamic background subtraction are evaluated to show the effectiveness of the proposed model, compared with conventional 2D-PCA and robust PCA algorithms.