In this paper, we present a novel deep-learning model for deformable registration of ultrasound images and an unsupervised approach to training this model. Our network employs recurrent all-pairs field transforms (RAFT) and a spatial transformer network (STN) to generate displacement fields at online rates (apprx. 30 Hz) and accurately track pixel movement. We call our approach unsupervised recurrent all-pairs field transforms (U-RAFT). In this work, we use U-RAFT to track pixels in a sequence of ultrasound images to cancel out respiratory motion in lung ultrasound images. We demonstrate our method on in-vivo porcine lung videos. We show a reduction of 76% in average pixel movement in the porcine dataset using respiratory motion compensation strategy. We believe U-RAFT is a promising tool for compensating different kinds of motions like respiration and heartbeat in ultrasound images of deformable tissue.
In addition to relevance, diversity is an important yet less studied performance metric of cross-modal image retrieval systems, which is critical to user experience. Existing solutions for diversity-aware image retrieval either explicitly post-process the raw retrieval results from standard retrieval systems or try to learn multi-vector representations of images to represent their diverse semantics. However, neither of them is good enough to balance relevance and diversity. On the one hand, standard retrieval systems are usually biased to common semantics and seldom exploit diversity-aware regularization in training, which makes it difficult to promote diversity by post-processing. On the other hand, multi-vector representation methods are not guaranteed to learn robust multiple projections. As a result, irrelevant images and images of rare or unique semantics may be projected inappropriately, which degrades the relevance and diversity of the results generated by some typical algorithms like top-k. To cope with these problems, this paper presents a new method called CoLT that tries to generate much more representative and robust representations for accurately classifying images. Specifically, CoLT first extracts semantics-aware image features by enhancing the preliminary representations of an existing one-to-one cross-modal system with semantics-aware contrastive learning. Then, a transformer-based token classifier is developed to subsume all the features into their corresponding categories. Finally, a post-processing algorithm is designed to retrieve images from each category to form the final retrieval result. Extensive experiments on two real-world datasets Div400 and Div150Cred show that CoLT can effectively boost diversity, and outperforms the existing methods as a whole (with a higher F1 score).
The objective of a style transfer is to maintain the content of an image while transferring the style of another image. However, conventional research on style transfer has a significant limitation in preserving facial landmarks, such as the eyes, nose, and mouth, which are crucial for maintaining the identity of the image. In Korean portraits, the majority of individuals wear "Gat", a type of headdress exclusively worn by men. Owing to its distinct characteristics from the hair in ID photos, transferring the "Gat" is challenging. To address this issue, this study proposes a deep learning network that can perform style transfer, including the "Gat", while preserving the identity of the face. Unlike existing style transfer approaches, the proposed method aims to preserve texture, costume, and the "Gat" on the style image. The Generative Adversarial Network forms the backbone of the proposed network. The color, texture, and intensity were extracted differently based on the characteristics of each block and layer of the pre-trained VGG-16, and only the necessary elements during training were preserved using a facial landmark mask. The head area was presented using the eyebrow area to transfer the "Gat". Furthermore, the identity of the face was retained, and style correlation was considered based on the Gram matrix. The proposed approach demonstrated superior transfer and preservation performance compared to previous studies.
Text-to-Image (T2I) diffusion models have achieved remarkable success in synthesizing high-quality images conditioned on text prompts. Recent methods have tried to replicate the success by either training text-to-video (T2V) models on a very large number of text-video pairs or adapting T2I models on text-video pairs independently. Although the latter is computationally less expensive, it still takes a significant amount of time for per-video adaption. To address this issue, we propose SAVE, a novel spectral-shift-aware adaptation framework, in which we fine-tune the spectral shift of the parameter space instead of the parameters themselves. Specifically, we take the spectral decomposition of the pre-trained T2I weights and only control the change in the corresponding singular values, i.e. spectral shift, while freezing the corresponding singular vectors. To avoid drastic drift from the original T2I weights, we introduce a spectral shift regularizer that confines the spectral shift to be more restricted for large singular values and more relaxed for small singular values. Since we are only dealing with spectral shifts, the proposed method reduces the adaptation time significantly (approx. 10 times) and has fewer resource constrains for training. Such attributes posit SAVE to be more suitable for real-world applications, e.g. editing undesirable content during video streaming. We validate the effectiveness of SAVE with an extensive experimental evaluation under different settings, e.g. style transfer, object replacement, privacy preservation, etc.
Model inversion attacks (MIAs) are aimed at recovering private data from a target model's training set, which poses a threat to the privacy of deep learning models. MIAs primarily focus on the white-box scenario where the attacker has full access to the structure and parameters of the target model. However, practical applications are black-box, it is not easy for adversaries to obtain model-related parameters, and various models only output predicted labels. Existing black-box MIAs primarily focused on designing the optimization strategy, and the generative model is only migrated from the GAN used in white-box MIA. Our research is the pioneering study of feasible attack models in label-only black-box scenarios, to the best of our knowledge. In this paper, we develop a novel method of MIA using the conditional diffusion model to recover the precise sample of the target without any extra optimization, as long as the target model outputs the label. Two primary techniques are introduced to execute the attack. Firstly, select an auxiliary dataset that is relevant to the target model task, and the labels predicted by the target model are used as conditions to guide the training process. Secondly, target labels and random standard normally distributed noise are input into the trained conditional diffusion model, generating target samples with pre-defined guidance strength. We then filter out the most robust and representative samples. Furthermore, we propose for the first time to use Learned Perceptual Image Patch Similarity (LPIPS) as one of the evaluation metrics for MIA, with systematic quantitative and qualitative evaluation in terms of attack accuracy, realism, and similarity. Experimental results show that this method can generate similar and accurate data to the target without optimization and outperforms generators of previous approaches in the label-only scenario.
Pre-trained Vision-Language Foundation Models utilizing extensive image-text paired data have demonstrated unprecedented image-text association capabilities, achieving remarkable results across various downstream tasks. A critical challenge is how to make use of existing large-scale pre-trained VLMs, which are trained on common objects, to perform the domain-specific transfer for accomplishing domain-related downstream tasks. In this paper, we propose a new framework that includes the Domain Foundation Model (DFM), bridging the gap between the General Foundation Model (GFM) and domain-specific downstream tasks. Moreover, we present an image-text paired dataset in the field of remote sensing (RS), RS5M, which has 5 million RS images with English descriptions. The dataset is obtained from filtering publicly available image-text paired datasets and captioning label-only RS datasets with pre-trained VLM. These constitute the first large-scale RS image-text paired dataset. Additionally, we tried several Parameter-Efficient Fine-Tuning methods on RS5M to implement the DFM. Experimental results show that our proposed dataset are highly effective for various tasks, improving upon the baseline by $8 \% \sim 16 \%$ in zero-shot classification tasks, and obtaining good results in both Vision-Language Retrieval and Semantic Localization tasks. Finally, we show successful results of training the RS Stable Diffusion model using the RS5M, uncovering more use cases of the dataset.
The perception of color is one of the most important aspects of human vision. From an evolutionary perspective, the accurate perception of color is crucial to distinguishing friend from foe, and food from fatal poison. As a result, humans have developed a keen sense of color and are able to detect subtle differences in appearance, while also robustly identifying colors across illumination and viewing conditions. In this paper, we shall briefly review methods for adapting traditional color appearance and difference models to complex image stimuli, and propose mechanisms to improve their performance. In particular, we find that applying contrast sensitivity functions and local adaptation rules in an edge-aware manner improves image difference predictions.
In recent years, differential privacy has seen significant advancements in image classification; however, its application to video activity recognition remains under-explored. This paper addresses the challenges of applying differential privacy to video activity recognition, which primarily stem from: (1) a discrepancy between the desired privacy level for entire videos and the nature of input data processed by contemporary video architectures, which are typically short, segmented clips; and (2) the complexity and sheer size of video datasets relative to those in image classification, which render traditional differential privacy methods inadequate. To tackle these issues, we propose Multi-Clip DP-SGD, a novel framework for enforcing video-level differential privacy through clip-based classification models. This method samples multiple clips from each video, averages their gradients, and applies gradient clipping in DP-SGD without incurring additional privacy loss. Moreover, we incorporate a parameter-efficient transfer learning strategy to make the model scalable for large-scale video datasets. Through extensive evaluations on the UCF-101 and HMDB-51 datasets, our approach exhibits impressive performance, achieving 81% accuracy with a privacy budget of epsilon=5 on UCF-101, marking a 76% improvement compared to a direct application of DP-SGD. Furthermore, we demonstrate that our transfer learning strategy is versatile and can enhance differentially private image classification across an array of datasets including CheXpert, ImageNet, CIFAR-10, and CIFAR-100.
In this technical report, we briefly introduce the solution of our team VIELab-HUST for coded target restoration through atmospheric turbulence in CVPR 2023 UG$^2$+ Track 2.2. In this task, we propose an efficient multi-stage framework to restore a high quality image from distorted frames. Specifically, each distorted frame is initially aligned using image registration to suppress geometric distortion. We subsequently select the sharpest set of registered frames by employing a frame selection approach based on image sharpness, and average them to produce an image that is largely free of geometric distortion, albeit with blurriness. A learning-based deblurring method is then applied to remove the residual blur in the averaged image. Finally, post-processing techniques are utilized to further enhance the quality of the output image. Our framework is capable of handling different kinds of coded target dataset provided in the final testing phase, and ranked 1st on the final leaderboard. Our code will be available at https://github.com/xsqhust/Turbulence_Removal.
While a large amount of work has focused on designing adversarial attacks against image classifiers, only a few methods exist to attack semantic segmentation models. We show that attacking segmentation models presents task-specific challenges, for which we propose novel solutions. Our final evaluation protocol outperforms existing methods, and shows that those can overestimate the robustness of the models. Additionally, so far adversarial training, the most successful way for obtaining robust image classifiers, could not be successfully applied to semantic segmentation. We argue that this is because the task to be learned is more challenging, and requires significantly higher computational effort than for image classification. As a remedy, we show that by taking advantage of recent advances in robust ImageNet classifiers, one can train adversarially robust segmentation models at limited computational cost by fine-tuning robust backbones.