CustomContrast: A Multilevel Contrastive Perspective For Subject-Driven Text-to-Image Customization

Subject-driven text-to-image (T2I) customization has drawn significant interest in academia and industry. This task enables pre-trained models to generate novel images based on unique subjects. Existing studies adopt a self-reconstructive perspective, focusing on capturing all details of a single image, which will misconstrue the specific image's irrelevant attributes (e.g., view, pose, and background) as the subject intrinsic attributes. This misconstruction leads to both overfitting or underfitting of irrelevant and intrinsic attributes of the subject, i.e., these attributes are over-represented or under-represented simultaneously, causing a trade-off between similarity and controllability. In this study, we argue an ideal subject representation can be achieved by a cross-differential perspective, i.e., decoupling subject intrinsic attributes from irrelevant attributes via contrastive learning, which allows the model to focus more on intrinsic attributes through intra-consistency (features of the same subject are spatially closer) and inter-distinctiveness (features of different subjects have distinguished differences). Specifically, we propose CustomContrast, a novel framework, which includes a Multilevel Contrastive Learning (MCL) paradigm and a Multimodal Feature Injection (MFI) Encoder. The MCL paradigm is used to extract intrinsic features of subjects from high-level semantics to low-level appearance through crossmodal semantic contrastive learning and multiscale appearance contrastive learning. To facilitate contrastive learning, we introduce the MFI encoder to capture cross-modal representations. Extensive experiments show the effectiveness of CustomContrast in subject similarity and text controllability.

RSTeller: Scaling Up Visual Language Modeling in Remote Sensing with Rich Linguistic Semantics from Openly Available Data and Large Language Models

Abundant, well-annotated multimodal data in remote sensing are pivotal for aligning complex visual remote sensing (RS) scenes with human language, enabling the development of specialized vision language models across diverse RS interpretation tasks. However, annotating RS images with rich linguistic semantics at scale demands expertise in RS and substantial human labor, making it costly and often impractical. In this study, we propose a workflow that leverages large language models (LLMs) to generate multimodal datasets with semantically rich captions at scale from plain OpenStreetMap (OSM) data for images sourced from the Google Earth Engine (GEE) platform. This approach facilitates the generation of paired remote sensing data and can be readily scaled up using openly available data. Within this framework, we present RSTeller, a multimodal dataset comprising over 1 million RS images, each accompanied by multiple descriptive captions. Extensive experiments demonstrate that RSTeller enhances the performance of multiple existing vision language models for RS scene understanding through continual pre-training. Our methodology significantly reduces the manual effort and expertise needed for annotating remote sensing imagery while democratizing access to high-quality annotated data. This advancement fosters progress in visual language modeling and encourages broader participation in remote sensing research and applications. The RSTeller dataset is available at https://github.com/SlytherinGe/RSTeller.

Verification of Geometric Robustness of Neural Networks via Piecewise Linear Approximation and Lipschitz Optimisation

We address the problem of verifying neural networks against geometric transformations of the input image, including rotation, scaling, shearing, and translation. The proposed method computes provably sound piecewise linear constraints for the pixel values by using sampling and linear approximations in combination with branch-and-bound Lipschitz optimisation. A feature of the method is that it obtains tighter over-approximations of the perturbation region than the present state-of-the-art. We report results from experiments on a comprehensive set of benchmarks. We show that our proposed implementation resolves more verification cases than present approaches while being more computationally efficient.

Training Overhead Ratio: A Practical Reliability Metric for Large Language Model Training Systems

Large Language Models (LLMs) are revolutionizing the AI industry with their superior capabilities. Training these models requires large-scale GPU clusters and significant computing time, leading to frequent failures that significantly increase training costs. Despite its significance, this field lacks a metric for evaluating reliability. In this work, we introduce a novel reliability metric called \emph{Training Overhead Ratio} (TOR) to evaluate the reliability of fault-tolerant LLM training systems. TOR is defined as the ratio of optimal training time to the observed training time of a system, serving as a practical tool for users to estimate the actual time required to train an LLM on a given system. Furthermore, our investigation identifies the key factor for enhancing reliability and present TOR equations for various types of failures encountered in practice.

Boosting Diffusion Model for Spectrogram Up-sampling in Text-to-speech: An Empirical Study

Scaling text-to-speech (TTS) with autoregressive language model (LM) to large-scale datasets by quantizing waveform into discrete speech tokens is making great progress to capture the diversity and expressiveness in human speech, but the speech reconstruction quality from discrete speech token is far from satisfaction depending on the compressed speech token compression ratio. Generative diffusion models trained with score-matching loss and continuous normalized flow trained with flow-matching loss have become prominent in generation of images as well as speech. LM based TTS systems usually quantize speech into discrete tokens and generate these tokens autoregressively, and finally use a diffusion model to up sample coarse-grained speech tokens into fine-grained codec features or mel-spectrograms before reconstructing into waveforms with vocoder, which has a high latency and is not realistic for real time speech applications. In this paper, we systematically investigate varied diffusion models for up sampling stage, which is the main bottleneck for streaming synthesis of LM and diffusion-based architecture, we present the model architecture, objective and subjective metrics to show quality and efficiency improvement.

PhysAvatar: Learning the Physics of Dressed 3D Avatars from Visual Observations

Modeling and rendering photorealistic avatars is of crucial importance in many applications. Existing methods that build a 3D avatar from visual observations, however, struggle to reconstruct clothed humans. We introduce PhysAvatar, a novel framework that combines inverse rendering with inverse physics to automatically estimate the shape and appearance of a human from multi-view video data along with the physical parameters of the fabric of their clothes. For this purpose, we adopt a mesh-aligned 4D Gaussian technique for spatio-temporal mesh tracking as well as a physically based inverse renderer to estimate the intrinsic material properties. PhysAvatar integrates a physics simulator to estimate the physical parameters of the garments using gradient-based optimization in a principled manner. These novel capabilities enable PhysAvatar to create high-quality novel-view renderings of avatars dressed in loose-fitting clothes under motions and lighting conditions not seen in the training data. This marks a significant advancement towards modeling photorealistic digital humans using physically based inverse rendering with physics in the loop. Our project website is at: https://qingqing-zhao.github.io/PhysAvatar

Multimodal Fusion Method with Spatiotemporal Sequences and Relationship Learning for Valence-Arousal Estimation

This paper presents our approach for the VA (Valence-Arousal) estimation task in the ABAW6 competition. We devised a comprehensive model by preprocessing video frames and audio segments to extract visual and audio features. Through the utilization of Temporal Convolutional Network (TCN) modules, we effectively captured the temporal and spatial correlations between these features. Subsequently, we employed a Transformer encoder structure to learn long-range dependencies, thereby enhancing the model's performance and generalization ability. Our method leverages a multimodal data fusion approach, integrating pre-trained audio and video backbones for feature extraction, followed by TCN-based spatiotemporal encoding and Transformer-based temporal information capture. Experimental results demonstrate the effectiveness of our approach, achieving competitive performance in VA estimation on the AffWild2 dataset.

Robust TOA-based Localization with Inaccurate Anchors for MANET

Accurate node localization is vital for mobile ad hoc networks (MANETs). Current methods like Time of Arrival (TOA) can estimate node positions using imprecise baseplates and achieve the Cram\'er-Rao lower bound (CRLB) accuracy. In multi-hop MANETs, some nodes lack direct links to base anchors, depending on neighbor nodes as dynamic anchors for chain localization. However, the dynamic nature of MANETs challenges TOA's robustness due to the availability and accuracy of base anchors, coupled with ranging errors. To address the issue of cascading positioning error divergence, we first derive the CRLB for any primary node in MANETs as a metric to tackle localization error in cascading scenarios. Second, we propose an advanced two-step TOA method based on CRLB which is able to approximate target node's CRLB with only local neighbor information. Finally, simulation results confirm the robustness of our algorithm, achieving CRLB-level accuracy for small ranging errors and maintaining precision for larger errors compared to existing TOA methods.

Inferring Hybrid Neural Fluid Fields from Videos

We study recovering fluid density and velocity from sparse multiview videos. Existing neural dynamic reconstruction methods predominantly rely on optical flows; therefore, they cannot accurately estimate the density and uncover the underlying velocity due to the inherent visual ambiguities of fluid velocity, as fluids are often shapeless and lack stable visual features. The challenge is further pronounced by the turbulent nature of fluid flows, which calls for properly designed fluid velocity representations. To address these challenges, we propose hybrid neural fluid fields (HyFluid), a neural approach to jointly infer fluid density and velocity fields. Specifically, to deal with visual ambiguities of fluid velocity, we introduce a set of physics-based losses that enforce inferring a physically plausible velocity field, which is divergence-free and drives the transport of density. To deal with the turbulent nature of fluid velocity, we design a hybrid neural velocity representation that includes a base neural velocity field that captures most irrotational energy and a vortex particle-based velocity that models residual turbulent velocity. We show that our method enables recovering vortical flow details. Our approach opens up possibilities for various learning and reconstruction applications centered around 3D incompressible flow, including fluid re-simulation and editing, future prediction, and neural dynamic scene composition. Project website: https://kovenyu.com/HyFluid/

Hardening RGB-D Object Recognition Systems against Adversarial Patch Attacks

RGB-D object recognition systems improve their predictive performances by fusing color and depth information, outperforming neural network architectures that rely solely on colors. While RGB-D systems are expected to be more robust to adversarial examples than RGB-only systems, they have also been proven to be highly vulnerable. Their robustness is similar even when the adversarial examples are generated by altering only the original images' colors. Different works highlighted the vulnerability of RGB-D systems; however, there is a lacking of technical explanations for this weakness. Hence, in our work, we bridge this gap by investigating the learned deep representation of RGB-D systems, discovering that color features make the function learned by the network more complex and, thus, more sensitive to small perturbations. To mitigate this problem, we propose a defense based on a detection mechanism that makes RGB-D systems more robust against adversarial examples. We empirically show that this defense improves the performances of RGB-D systems against adversarial examples even when they are computed ad-hoc to circumvent this detection mechanism, and that is also more effective than adversarial training.