The field of vision and language has witnessed a proliferation of pre-trained foundation models. Most existing methods are independently pre-trained with contrastive objective like CLIP, image-to-text generative objective like PaLI, or text-to-image generative objective like Parti. However, the three objectives can be pre-trained on the same data, image-text pairs, and intuitively they complement each other as contrasting provides global alignment capacity and generation grants fine-grained understanding. In this work, we present a Contrastive Bi-directional Image-Text generation model (CoBIT), which attempts to unify the three pre-training objectives in one framework. Specifically, CoBIT employs a novel unicoder-decoder structure, consisting of an image unicoder, a text unicoder and a cross-modal decoder. The image/text unicoders can switch between encoding and decoding in different tasks, enabling flexibility and shared knowledge that benefits both image-to-text and text-to-image generations. CoBIT achieves superior performance in image understanding, image-text understanding (Retrieval, Captioning, VQA, SNLI-VE) and text-based content creation, particularly in zero-shot scenarios. For instance, 82.7% in zero-shot ImageNet classification, 9.37 FID score in zero-shot text-to-image generation and 44.8 CIDEr in zero-shot captioning.
Deep image classification models trained on large amounts of web-scraped data are vulnerable to data poisoning, a mechanism for backdooring models. Even a few poisoned samples seen during training can entirely undermine the model's integrity during inference. While it is known that poisoning more samples enhances an attack's effectiveness and robustness, it is unknown whether poisoning too many samples weakens an attack by making it more detectable. We observe a fundamental detectability/robustness trade-off in data poisoning attacks: Poisoning too few samples renders an attack ineffective and not robust, but poisoning too many samples makes it detectable. This raises the bar for data poisoning attackers who have to balance this trade-off to remain robust and undetectable. Our work proposes two defenses designed to (i) detect and (ii) repair poisoned models as a post-processing step after training using a limited amount of trusted image-label pairs. We show that our defenses mitigate all surveyed attacks and outperform existing defenses using less trusted data to repair a model. Our defense scales to joint vision-language models, such as CLIP, and interestingly, we find that attacks on larger models are more easily detectable but also more robust than those on smaller models. Lastly, we propose two adaptive attacks demonstrating that while our work raises the bar for data poisoning attacks, it cannot mitigate all forms of backdooring.
Despite its success in image synthesis, we observe that diffusion probabilistic models (DPMs) often lack contextual reasoning ability to learn the relations among object parts in an image, leading to a slow learning process. To solve this issue, we propose a Masked Diffusion Transformer (MDT) that introduces a mask latent modeling scheme to explicitly enhance the DPMs' ability of contextual relation learning among object semantic parts in an image. During training, MDT operates on the latent space to mask certain tokens. Then, an asymmetric masking diffusion transformer is designed to predict masked tokens from unmasked ones while maintaining the diffusion generation process. Our MDT can reconstruct the full information of an image from its incomplete contextual input, thus enabling it to learn the associated relations among image tokens. Experimental results show that MDT achieves superior image synthesis performance, e.g. a new SoTA FID score on the ImageNet dataset, and has about 3x faster learning speed than the previous SoTA DiT. The source code is released at https://github.com/sail-sg/MDT.
Conditioning image generation on specific features of the desired output is a key ingredient of modern generative models. Most existing approaches focus on conditioning the generation based on free-form text, while some niche studies use scene graphs to describe the content of the image to be generated. This paper explores novel methods to condition image generation that are based on object-centric relational representations. In particular, we propose a methodology to condition the generation of a particular object in an image on the attributed graph representing its structure and associated style. We show that such architectural biases entail properties that facilitate the manipulation and conditioning of the generative process and allow for regularizing the training procedure. The proposed framework is implemented by means of a neural network architecture combining convolutional operators that operate on both the underlying graph and the 2D grid that becomes the output image. The resulting model learns to generate multi-channel masks of the object that can be used as a soft inductive bias in the downstream generative task. Empirical results show that the proposed approach compares favorably against relevant baselines on image generation conditioned on human poses.
The local beat-to-beat local pulse pressure (PP) and blood pressure waveform of arteries, especially central arteries, are important indicators of the course of cardiovascular diseases (CVDs). Nevertheless, noninvasive measurement of them remains a challenge in the clinic. This work presents a three-element image-free ultrasound system with a low-computational method for real-time measurement of local pulse wave velocity (PWV) and diameter waveforms, enabling real-time and noninvasive continuous PP and blood pressure waveforms measurement without calibration. The developed system has been well-validated in vitro and in vivo. In in vitro cardiovascular phantom experiments, the results demonstrated high accuracy in the measurement of PP (error < 3 mmHg) and blood pressure waveform (root-mean-square-errors (RMSE) < 2 mmHg, correlation coefficient (r) > textgreater 0.99). In subsequent human carotid experiments, the system was compared with an arterial tonometer, which showed excellent PP accuracy (mean absolute error (MAE) = 3.7 +- 3.4 mmHg) and pressure waveform similarity (RMSE = 3.7 +- 1.6 mmHg, r = 0.98 +- 0.01). Furthermore, comparative experiments with the volume clamp device demonstrated the system's ability to accurately trace blood pressure changes (induced by deep breathing) over a period of one minute, with the MAE of DBP, MAP, and SBP within 5 +- 8 mmHg. The present results demonstrate the accuracy and reliability of the developed system for continuous and noninvasive measurement of arterial PP and blood pressure waveform measurements, with potential applications in the diagnosis and prevention of CVDs.
We present the unified computational framework for modeling the sulcal patterns of human brain obtained from the magnetic resonance images. The Wasserstein distance is used to align the sulcal patterns nonlinearly. These patterns are topologically different across subjects making the pattern matching a challenge. We work out the mathematical details and develop the gradient descent algorithms for estimating the deformation field. We further quantify the image registration performance. This method is applied in identifying the differences between male and female sulcal patterns.
Multi-view image generation attracts particular attention these days due to its promising 3D-related applications, e.g., image viewpoint editing. Most existing methods follow a paradigm where a 3D representation is first synthesized, and then rendered into 2D images to ensure photo-consistency across viewpoints. However, such explicit bias for photo-consistency sacrifices photo-realism, causing geometry artifacts and loss of fine-scale details when these methods are applied to edit real images. To address this issue, we propose ray conditioning, a geometry-free alternative that relaxes the photo-consistency constraint. Our method generates multi-view images by conditioning a 2D GAN on a light field prior. With explicit viewpoint control, state-of-the-art photo-realism and identity consistency, our method is particularly suited for the viewpoint editing task.
Modern image inpainting systems, despite the significant progress, often struggle with mask selection and holes filling. Based on Segment-Anything Model (SAM), we make the first attempt to the mask-free image inpainting and propose a new paradigm of ``clicking and filling'', which is named as Inpaint Anything (IA). The core idea behind IA is to combine the strengths of different models in order to build a very powerful and user-friendly pipeline for solving inpainting-related problems. IA supports three main features: (i) Remove Anything: users could click on an object and IA will remove it and smooth the ``hole'' with the context; (ii) Fill Anything: after certain objects removal, users could provide text-based prompts to IA, and then it will fill the hole with the corresponding generative content via driving AIGC models like Stable Diffusion; (iii) Replace Anything: with IA, users have another option to retain the click-selected object and replace the remaining background with the newly generated scenes. We are also very willing to help everyone share and promote new projects based on our Inpaint Anything (IA). Our codes are available at https://github.com/geekyutao/Inpaint-Anything.
Diplomatics, the analysis of medieval charters, is a major field of research in which paleography is applied. Annotating data, if performed by laymen, needs validation and correction by experts. In this paper, we propose an effective and efficient annotation approach for charter segmentation, essentially reducing it to object detection. This approach allows for a much more efficient use of the paleographer's time and produces results that can compete and even outperform pixel-level segmentation in some use cases. Further experiments shed light on how to design a class ontology in order to make the best use of annotators' time and effort. Exploiting the presence of calibration cards in the image, we further annotate the data with the physical length in pixels and train regression neural networks to predict it from image patches.
Recent text-to-image generation models like DreamBooth have made remarkable progress in generating highly customized images of a target subject, by fine-tuning an ``expert model'' for a given subject from a few examples. However, this process is expensive, since a new expert model must be learned for each subject. In this paper, we present SuTI, a Subject-driven Text-to-Image generator that replaces subject-specific fine tuning with \emph{in-context} learning. Given a few demonstrations of a new subject, SuTI can instantly generate novel renditions of the subject in different scenes, without any subject-specific optimization. SuTI is powered by {\em apprenticeship learning}, where a single apprentice model is learned from data generated by massive amount of subject-specific expert models. Specifically, we mine millions of image clusters from the Internet, each centered around a specific visual subject. We adopt these clusters to train massive amount of expert models specialized on different subjects. The apprentice model SuTI then learns to mimic the behavior of these experts through the proposed apprenticeship learning algorithm. SuTI can generate high-quality and customized subject-specific images 20x faster than optimization-based SoTA methods. On the challenging DreamBench and DreamBench-v2, our human evaluation shows that SuTI can significantly outperform existing approaches like InstructPix2Pix, Textual Inversion, Imagic, Prompt2Prompt, Re-Imagen while performing on par with DreamBooth.