Recent years have witnessed the strong power of large text-to-image diffusion models for the impressive generative capability to create high-fidelity images. However, it is very tricky to generate desired images using only text prompt as it often involves complex prompt engineering. An alternative to text prompt is image prompt, as the saying goes: "an image is worth a thousand words". Although existing methods of direct fine-tuning from pretrained models are effective, they require large computing resources and are not compatible with other base models, text prompt, and structural controls. In this paper, we present IP-Adapter, an effective and lightweight adapter to achieve image prompt capability for the pretrained text-to-image diffusion models. The key design of our IP-Adapter is decoupled cross-attention mechanism that separates cross-attention layers for text features and image features. Despite the simplicity of our method, an IP-Adapter with only 22M parameters can achieve comparable or even better performance to a fully fine-tuned image prompt model. As we freeze the pretrained diffusion model, the proposed IP-Adapter can be generalized not only to other custom models fine-tuned from the same base model, but also to controllable generation using existing controllable tools. With the benefit of the decoupled cross-attention strategy, the image prompt can also work well with the text prompt to achieve multimodal image generation. The project page is available at \url{//ip-adapter.github.io}.
相關內容
In Ultrasound Localization Microscopy (ULM), achieving high-resolution images relies on the precise localization of contrast agent particles across consecutive beamformed frames. However, our study uncovers an enormous potential: The process of delay-and-sum beamforming leads to an irreversible reduction of Radio-Frequency (RF) data, while its implications for localization remain largely unexplored. The rich contextual information embedded within RF wavefronts, including their hyperbolic shape and phase, offers great promise for guiding Deep Neural Networks (DNNs) in challenging localization scenarios. To fully exploit this data, we propose to directly localize scatterers in RF signals. Our approach involves a custom super-resolution DNN using learned feature channel shuffling and a novel semi-global convolutional sampling block tailored for reliable and accurate localization in RF input data. Additionally, we introduce a geometric point transformation that facilitates seamless mapping between B-mode and RF spaces. To validate the effectiveness of our method and understand the impact of beamforming, we conduct an extensive comparison with State-Of-The-Art (SOTA) techniques in ULM. We present the inaugural in vivo results from an RF-trained DNN, highlighting its real-world practicality. Our findings show that RF-ULM bridges the domain gap between synthetic and real datasets, offering a considerable advantage in terms of precision and complexity. To enable the broader research community to benefit from our findings, our code and the associated SOTA methods are made available at //github.com/hahnec/rf-ulm.
H-InDex: Visual Reinforcement Learning with Hand-Informed Representations for Dexterous Manipulation
Human hands possess remarkable dexterity and have long served as a source of inspiration for robotic manipulation. In this work, we propose a human $\textbf{H}$and$\textbf{-In}$formed visual representation learning framework to solve difficult $\textbf{Dex}$terous manipulation tasks ($\textbf{H-InDex}$) with reinforcement learning. Our framework consists of three stages: (i) pre-training representations with 3D human hand pose estimation, (ii) offline adapting representations with self-supervised keypoint detection, and (iii) reinforcement learning with exponential moving average BatchNorm. The last two stages only modify $0.36\%$ parameters of the pre-trained representation in total, ensuring the knowledge from pre-training is maintained to the full extent. We empirically study 12 challenging dexterous manipulation tasks and find that H-InDex largely surpasses strong baseline methods and the recent visual foundation models for motor control. Code is available at //yanjieze.com/H-InDex .
Deep neural networks are vulnerable to backdoor attacks, where an adversary maliciously manipulates the model behavior through overlaying images with special triggers. Existing backdoor defense methods often require accessing a few validation data and model parameters, which are impractical in many real-world applications, e.g., when the model is provided as a cloud service. In this paper, we address the practical task of blind backdoor defense at test time, in particular for black-box models. The true label of every test image needs to be recovered on the fly from a suspicious model regardless of image benignity. We focus on test-time image purification methods that incapacitate possible triggers while keeping semantic contents intact. Due to diverse trigger patterns and sizes, the heuristic trigger search in image space can be unscalable. We circumvent such barrier by leveraging the strong reconstruction power of generative models, and propose a framework of Blind Defense with Masked AutoEncoder (BDMAE). It detects possible triggers in the token space using image structural similarity and label consistency between the test image and MAE restorations. The detection results are then refined by considering trigger topology. Finally, we fuse MAE restorations adaptively into a purified image for making prediction. Our approach is blind to the model architectures, trigger patterns and image benignity. Extensive experiments under different backdoor settings validate its effectiveness and generalizability. Code is available at //github.com/tsun/BDMAE.
We explore the task of embodied view synthesis from monocular videos of deformable scenes. Given a minute-long RGBD video of people interacting with their pets, we render the scene from novel camera trajectories derived from the in-scene motion of actors: (1) egocentric cameras that simulate the point of view of a target actor and (2) 3rd-person cameras that follow the actor. Building such a system requires reconstructing the root-body and articulated motion of every actor, as well as a scene representation that supports free-viewpoint synthesis. Longer videos are more likely to capture the scene from diverse viewpoints (which helps reconstruction) but are also more likely to contain larger motions (which complicates reconstruction). To address these challenges, we present Total-Recon, the first method to photorealistically reconstruct deformable scenes from long monocular RGBD videos. Crucially, to scale to long videos, our method hierarchically decomposes the scene into the background and objects, whose motion is decomposed into carefully initialized root-body motion and local articulations. To quantify such "in-the-wild" reconstruction and view synthesis, we collect ground-truth data from a specialized stereo RGBD capture rig for 11 challenging videos, significantly outperforming prior methods. Our code, model, and data can be found at //andrewsonga.github.io/totalrecon .
Lossy image coding standards such as JPEG and MPEG have successfully achieved high compression rates for human consumption of multimedia data. However, with the increasing prevalence of IoT devices, drones, and self-driving cars, machines rather than humans are processing a greater portion of captured visual content. Consequently, it is crucial to pursue an efficient compressed representation that caters not only to human vision but also to image processing and machine vision tasks. Drawing inspiration from the efficient coding hypothesis in biological systems and the modeling of the sensory cortex in neural science, we repurpose the compressed latent representation to prioritize semantic relevance while preserving perceptual distance. Our proposed method, Compressed Perceptual Image Patch Similarity (CPIPS), can be derived at a minimal cost from a learned neural codec and computed significantly faster than DNN-based perceptual metrics such as LPIPS and DISTS.
Most of the existing multi-modal models, hindered by their incapacity to adeptly manage interleaved image-and-text inputs in multi-image, multi-round dialogues, face substantial constraints in resource allocation for training and data accessibility, impacting their adaptability and scalability across varied interaction realms. To address this, we present the DeepSpeed-VisualChat framework, designed to optimize Large Language Models (LLMs) by incorporating multi-modal capabilities, with a focus on enhancing the proficiency of Large Vision and Language Models in handling interleaved inputs. Our framework is notable for (1) its open-source support for multi-round and multi-image dialogues, (2) introducing an innovative multi-modal causal attention mechanism, and (3) utilizing data blending techniques on existing datasets to assure seamless interactions in multi-round, multi-image conversations. Compared to existing frameworks, DeepSpeed-VisualChat shows superior scalability up to 70B parameter language model size, representing a significant advancement in multi-modal language models and setting a solid foundation for future explorations.
The current Internet lacks a bandwidth-reservation infrastructure that enables fine-grained inter-domain reservations for end hosts. This is hindering the provisioning of quality-of-service guarantees for real-time applications like video calls and gaming, cloud-based systems, financial transactions, telesurgery, and other remote applications that benefit from reliable communication. This paper introduces Hummingbird, a novel lightweight inter-domain bandwidth-reservation system that addresses several shortcomings of previous designs. Hummingbird supports flexible and composable reservations and enables end-to-end guarantees without requiring autonomous systems to manage reservations for their endhosts. Previous systems tied reservations to autonomous-system numbers or network addresses, which limits the flexibility of reservations. In contrast, our system decouples reservations from network identities and, as a result, the control plane from the data plane. This design choice facilitates multiple co-existing control-plane mechanisms and enables innovative approaches, such as a control plane based on blockchain smart contracts that offers tradeable bandwidth-reservation assets and end-to-end guarantees. The data-plane design ensures simplicity for efficient processing on border routers, which streamlines implementation, deployment, and traffic policing while maintaining robust security properties.
We propose a neuralized undirected graphical model called Neural-Hidden-CRF to solve the weakly-supervised sequence labeling problem. Under the umbrella of probabilistic undirected graph theory, the proposed Neural-Hidden-CRF embedded with a hidden CRF layer models the variables of word sequence, latent ground truth sequence, and weak label sequence with the global perspective that undirected graphical models particularly enjoy. In Neural-Hidden-CRF, we can capitalize on the powerful language model BERT or other deep models to provide rich contextual semantic knowledge to the latent ground truth sequence, and use the hidden CRF layer to capture the internal label dependencies. Neural-Hidden-CRF is conceptually simple and empirically powerful. It obtains new state-of-the-art results on one crowdsourcing benchmark and three weak-supervision benchmarks, including outperforming the recent advanced model CHMM by 2.80 F1 points and 2.23 F1 points in average generalization and inference performance, respectively.
Hierarchical structures are popular in recent vision transformers, however, they require sophisticated designs and massive datasets to work well. In this paper, we explore the idea of nesting basic local transformers on non-overlapping image blocks and aggregating them in a hierarchical way. We find that the block aggregation function plays a critical role in enabling cross-block non-local information communication. This observation leads us to design a simplified architecture that requires minor code changes upon the original vision transformer. The benefits of the proposed judiciously-selected design are threefold: (1) NesT converges faster and requires much less training data to achieve good generalization on both ImageNet and small datasets like CIFAR; (2) when extending our key ideas to image generation, NesT leads to a strong decoder that is 8$\times$ faster than previous transformer-based generators; and (3) we show that decoupling the feature learning and abstraction processes via this nested hierarchy in our design enables constructing a novel method (named GradCAT) for visually interpreting the learned model. Source code is available //github.com/google-research/nested-transformer.
Answering complex questions about images is an ambitious goal for machine intelligence, which requires a joint understanding of images, text, and commonsense knowledge, as well as a strong reasoning ability. Recently, multimodal Transformers have made great progress in the task of Visual Commonsense Reasoning (VCR), by jointly understanding visual objects and text tokens through layers of cross-modality attention. However, these approaches do not utilize the rich structure of the scene and the interactions between objects which are essential in answering complex commonsense questions. We propose a Scene Graph Enhanced Image-Text Learning (SGEITL) framework to incorporate visual scene graphs in commonsense reasoning. To exploit the scene graph structure, at the model structure level, we propose a multihop graph transformer for regularizing attention interaction among hops. As for pre-training, a scene-graph-aware pre-training method is proposed to leverage structure knowledge extracted in the visual scene graph. Moreover, we introduce a method to train and generate domain-relevant visual scene graphs using textual annotations in a weakly-supervised manner. Extensive experiments on VCR and other tasks show a significant performance boost compared with the state-of-the-art methods and prove the efficacy of each proposed component.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191