Diffusion probabilistic models (DPMs) have shown remarkable results on various image synthesis tasks such as text-to-image generation and image inpainting. However, compared to other generative methods like VAEs and GANs, DPMs lack a low-dimensional, interpretable, and well-decoupled latent code. Recently, diffusion autoencoders (Diff-AE) were proposed to explore the potential of DPMs for representation learning via autoencoding. Diff-AE provides an accessible latent space that exhibits remarkable interpretability, allowing us to manipulate image attributes based on latent codes from the space. However, previous works are not generic as they only operated on a few limited attributes. To further explore the latent space of Diff-AE and achieve a generic editing pipeline, we proposed a module called Group-supervised AutoEncoder(dubbed GAE) for Diff-AE to achieve better disentanglement on the latent code. Our proposed GAE has trained via an attribute-swap strategy to acquire the latent codes for multi-attribute image manipulation based on examples. We empirically demonstrate that our method enables multiple-attributes manipulation and achieves convincing sample quality and attribute alignments, while significantly reducing computational requirements compared to pixel-based approaches for representational decoupling. Code will be released soon.
Graph-based neural network models are gaining traction in the field of representation learning due to their ability to uncover latent topological relationships between entities that are otherwise challenging to identify. These models have been employed across a diverse range of domains, encompassing drug discovery, protein interactions, semantic segmentation, and fluid dynamics research. In this study, we investigate the potential of Graph Neural Networks (GNNs) for medical image classification. We introduce a novel model that combines GNNs and edge convolution, leveraging the interconnectedness of RGB channel feature values to strongly represent connections between crucial graph nodes. Our proposed model not only performs on par with state-of-the-art Deep Neural Networks (DNNs) but does so with 1000 times fewer parameters, resulting in reduced training time and data requirements. We compare our Graph Convolutional Neural Network (GCNN) to pre-trained DNNs for classifying MedMNIST dataset classes, revealing promising prospects for GNNs in medical image analysis. Our results also encourage further exploration of advanced graph-based models such as Graph Attention Networks (GAT) and Graph Auto-Encoders in the medical imaging domain. The proposed model yields more reliable, interpretable, and accurate outcomes for tasks like semantic segmentation and image classification compared to simpler GCNNs
Generative text-to-image models such as Stable Diffusion allow users to generate images based on a textual description, the prompt. Changing the prompt is still the primary means for the user to change a generated image as desired. However, changing the image by reformulating the prompt remains a difficult process of trial and error, which has led to the emergence of prompt engineering as a new field of research. We propose and analyze methods to change the embedding of a prompt directly instead of the prompt text. It allows for more fine-grained and targeted control that takes into account user intentions. Our approach treats the generative text-to-image model as a continuous function and passes gradients between the image space and the prompt embedding space. By addressing different user interaction problems, we can apply this idea in three scenarios: (1) Optimization of a metric defined in image space that could measure, for example, image style. (2) Assistance of users in creative tasks by enabling them to navigate the image space along a selection of directions of "near" prompt embeddings. (3) Changing the embedding of the prompt to include information that the user has seen in a particular seed but finds difficult to describe in the prompt. Our experiments demonstrate the feasibility of the described methods.
Unpaired Medical Image Enhancement (UMIE) aims to transform a low-quality (LQ) medical image into a high-quality (HQ) one without relying on paired images for training. While most existing approaches are based on Pix2Pix/CycleGAN and are effective to some extent, they fail to explicitly use HQ information to guide the enhancement process, which can lead to undesired artifacts and structural distortions. In this paper, we propose a novel UMIE approach that avoids the above limitation of existing methods by directly encoding HQ cues into the LQ enhancement process in a variational fashion and thus model the UMIE task under the joint distribution between the LQ and HQ domains. Specifically, we extract features from an HQ image and explicitly insert the features, which are expected to encode HQ cues, into the enhancement network to guide the LQ enhancement with the variational normalization module. We train the enhancement network adversarially with a discriminator to ensure the generated HQ image falls into the HQ domain. We further propose a content-aware loss to guide the enhancement process with wavelet-based pixel-level and multi-encoder-based feature-level constraints. Additionally, as a key motivation for performing image enhancement is to make the enhanced images serve better for downstream tasks, we propose a bi-level learning scheme to optimize the UMIE task and downstream tasks cooperatively, helping generate HQ images both visually appealing and favorable for downstream tasks. Experiments on three medical datasets, including two newly collected datasets, verify that the proposed method outperforms existing techniques in terms of both enhancement quality and downstream task performance. We will make the code and the newly collected datasets publicly available for community study.
Large-scale Pre-Training Vision-Language Model such as CLIP has demonstrated outstanding performance in zero-shot classification, e.g. achieving 76.3% top-1 accuracy on ImageNet without seeing any example, which leads to potential benefits to many tasks that have no labeled data. However, while applying CLIP to a downstream target domain, the presence of visual and text domain gaps and cross-modality misalignment can greatly impact the model performance. To address such challenges, we propose ReCLIP, the first source-free domain adaptation method for vision-language models, which does not require any source data or target labeled data. ReCLIP first learns a projection space to mitigate the misaligned visual-text embeddings and learns pseudo labels, and then deploys cross-modality self-training with the pseudo labels, to update visual and text encoders, refine labels and reduce domain gaps and misalignments iteratively. With extensive experiments, we demonstrate ReCLIP reduces the average error rate of CLIP from 30.17% to 25.06% on 22 image classification benchmarks.
Objective. A phased or a curvilinear array produces ultrasound (US) images with a sector field of view (FOV), which inherently exhibits spatially-varying image resolution with inferior quality in the far zone and towards the two sides azimuthally. Sector US images with improved spatial resolutions are favorable for accurate quantitative analysis of large and dynamic organs, such as the heart. Therefore, this study aims to translate US images with spatially-varying resolution to ones with less spatially-varying resolution. CycleGAN has been a prominent choice for unpaired medical image translation; however, it neither guarantees structural consistency nor preserves backscattering patterns between input and generated images for unpaired US images. Approach. To circumvent this limitation, we propose a constrained CycleGAN (CCycleGAN), which directly performs US image generation with unpaired images acquired by different ultrasound array probes. In addition to conventional adversarial and cycle-consistency losses of CycleGAN, CCycleGAN introduces an identical loss and a correlation coefficient loss based on intrinsic US backscattered signal properties to constrain structural consistency and backscattering patterns, respectively. Instead of post-processed B-mode images, CCycleGAN uses envelope data directly obtained from beamformed radio-frequency signals without any other non-linear postprocessing. Main Results. In vitro phantom results demonstrate that CCycleGAN successfully generates images with improved spatial resolution as well as higher peak signal-to-noise ratio (PSNR) and structural similarity (SSIM) compared with benchmarks. Significance. CCycleGAN-generated US images of the in vivo human beating heart further facilitate higher quality heart wall motion estimation than benchmarks-generated ones, particularly in deep regions.
Due to the large-scale availability of data, machine learning (ML) algorithms are being deployed in distributed topologies, where different nodes collaborate to train ML models over their individual data by exchanging model-related information (e.g., gradients) with a central server. However, distributed learning schemes are notably vulnerable to two threats. First, Byzantine nodes can single-handedly corrupt the learning by sending incorrect information to the server, e.g., erroneous gradients. The standard approach to mitigate such behavior is to use a non-linear robust aggregation method at the server. Second, the server can violate the privacy of the nodes. Recent attacks have shown that exchanging (unencrypted) gradients enables a curious server to recover the totality of the nodes' data. The use of homomorphic encryption (HE), a gold standard security primitive, has extensively been studied as a privacy-preserving solution to distributed learning in non-Byzantine scenarios. However, due to HE's large computational demand especially for high-dimensional ML models, there has not yet been any attempt to design purely homomorphic operators for non-linear robust aggregators. In this work, we present SABLE, the first completely homomorphic and Byzantine robust distributed learning algorithm. SABLE essentially relies on a novel plaintext encoding method that enables us to implement the robust aggregator over batching-friendly BGV. Moreover, this encoding scheme also accelerates state-of-the-art homomorphic sorting with larger security margins and smaller ciphertext size. We perform extensive experiments on image classification tasks and show that our algorithm achieves practical execution times while matching the ML performance of its non-private counterpart.
A holistic understanding of object properties across diverse sensory modalities (e.g., visual, audio, and haptic) is essential for tasks ranging from object categorization to complex manipulation. Drawing inspiration from cognitive science studies that emphasize the significance of multi-sensory integration in human perception, we introduce MOSAIC (Multi-modal Object property learning with Self-Attention and Integrated Comprehension), a novel framework designed to facilitate the learning of unified multi-sensory object property representations. While it is undeniable that visual information plays a prominent role, we acknowledge that many fundamental object properties extend beyond the visual domain to encompass attributes like texture, mass distribution, or sounds, which significantly influence how we interact with objects. In MOSAIC, we leverage this profound insight by distilling knowledge from the extensive pre-trained Contrastive Language-Image Pre-training (CLIP) model, aligning these representations not only across vision but also haptic and auditory sensory modalities. Through extensive experiments on a dataset where a humanoid robot interacts with 100 objects across 10 exploratory behaviors, we demonstrate the versatility of MOSAIC in two task families: object categorization and object-fetching tasks. Our results underscore the efficacy of MOSAIC's unified representations, showing competitive performance in category recognition through a simple linear probe setup and excelling in the fetch object task under zero-shot transfer conditions. This work pioneers the application of CLIP-based sensory grounding in robotics, promising a significant leap in multi-sensory perception capabilities for autonomous systems. We have released the code, datasets, and additional results: https://github.com/gtatiya/MOSAIC.
Multispectral Optoacoustic Tomography (MSOT) uniquely enables spatial mapping in high resolution of oxygen saturation (SO$_2$), with potential applications in studying pathological complications and therapy efficacy. MSOT offers seamless integration with ultrasonography, by using a common ultrasound detector array. However, MSOT relies on multiple successive acquisitions of optoacoustic (OA) images at different optical wavelengths and the low frame rate of OA imaging makes the MSOT acquisition sensitive to body/respiratory motion. Moreover, estimation of SO$_2$ is highly sensitive to noise, and artefacts related to the respiratory motion of the animal were identified as the primary source of noise in MSOT.In this work, we propose a two-step image processing method for SO$_2$ estimation in deep tissues. First, to mitigate motion artefacts, we propose a method of selection of OA images acquired only during the respiratory pause of the animal, using ultrafast ultrasound images (USIs) acquired immediately after each OA acquisition (USI acquisition duration of 1.4 ms and a total delay of 7 ms). We show that gating is more effective using USIs than OA images at different optical wavelengths. Secondly, we propose a novel method which can estimate directly the SO$_2$ value of a pixel and at the same time evaluate the amount of noise present in that pixel. Hence, the method can efficiently eliminate the pixels dominated by noise from the final SO$_2$ map. Our post-processing method is shown to outperform conventional methods for SO$_2$ estimation, and the method was validated by in vivo oxygen challenge experiments.
Recent advances in vision-language models (VLMs) have led to improved performance on tasks such as visual question answering and image captioning. Consequently, these models are now well-positioned to reason about the physical world, particularly within domains such as robotic manipulation. However, current VLMs are limited in their understanding of the physical concepts (e.g., material, fragility) of common objects, which restricts their usefulness for robotic manipulation tasks that involve interaction and physical reasoning about such objects. To address this limitation, we propose PhysObjects, an object-centric dataset of 39.6K crowd-sourced and 417K automated physical concept annotations of common household objects. We demonstrate that fine-tuning a VLM on PhysObjects improves its understanding of physical object concepts, including generalization to held-out concepts, by capturing human priors of these concepts from visual appearance. We incorporate this physically-grounded VLM in an interactive framework with a large language model-based robotic planner, and show improved planning performance on tasks that require reasoning about physical object concepts, compared to baselines that do not leverage physically-grounded VLMs. We additionally illustrate the benefits of our physically-grounded VLM on a real robot, where it improves task success rates. We release our dataset and provide further details and visualizations of our results at https://iliad.stanford.edu/pg-vlm/.