Abstract:Mamba has demonstrated exceptional performance in visual tasks due to its powerful global modeling capabilities and linear computational complexity, offering considerable potential in hyperspectral image super-resolution (HSISR). However, in HSISR, Mamba faces challenges as transforming images into 1D sequences neglects the spatial-spectral structural relationships between locally adjacent pixels, and its performance is highly sensitive to input order, which affects the restoration of both spatial and spectral details. In this paper, we propose HSRMamba, a contextual spatial-spectral modeling state space model for HSISR, to address these issues both locally and globally. Specifically, a local spatial-spectral partitioning mechanism is designed to establish patch-wise causal relationships among adjacent pixels in 3D features, mitigating the local forgetting issue. Furthermore, a global spectral reordering strategy based on spectral similarity is employed to enhance the causal representation of similar pixels across both spatial and spectral dimensions. Finally, experimental results demonstrate our HSRMamba outperforms the state-of-the-art methods in quantitative quality and visual results. Code will be available soon.
Abstract:Cushing's syndrome is a condition caused by excessive glucocorticoid secretion from the adrenal cortex, often manifesting with moon facies and plethora, making facial data crucial for diagnosis. Previous studies have used pre-trained convolutional neural networks (CNNs) for diagnosing Cushing's syndrome using frontal facial images. However, CNNs are better at capturing local features, while Cushing's syndrome often presents with global facial features. Transformer-based models like ViT and SWIN, which utilize self-attention mechanisms, can better capture long-range dependencies and global features. Recently, DINOv2, a foundation model based on visual Transformers, has gained interest. This study compares the performance of various pre-trained models, including CNNs, Transformer-based models, and DINOv2, in diagnosing Cushing's syndrome. We also analyze gender bias and the impact of freezing mechanisms on DINOv2. Our results show that Transformer-based models and DINOv2 outperformed CNNs, with ViT achieving the highest F1 score of 85.74%. Both the pre-trained model and DINOv2 had higher accuracy for female samples. DINOv2 also showed improved performance when freezing parameters. In conclusion, Transformer-based models and DINOv2 are effective for Cushing's syndrome classification.
Abstract:The growing availability of sensors within semiconductor manufacturing processes makes it feasible to detect defective wafers with data-driven models. Without directly measuring the quality of semiconductor devices, they capture the modalities between diverse sensor readings and can be used to predict key quality indicators (KQI, \textit{e.g.}, roughness, resistance) to detect faulty products, significantly reducing the capital and human cost in maintaining physical metrology steps. Nevertheless, existing models pay little attention to the correlations among different processes for diverse wafer products and commonly struggle with generalizability issues. To enable generic fault detection, in this work, we propose a modular network (MN) trained using time series stage-wise datasets that embodies the structure of the manufacturing process. It decomposes KQI prediction as a combination of stage modules to simulate compositional semiconductor manufacturing, universally enhancing faulty wafer detection among different wafer types and manufacturing processes. Extensive experiments demonstrate the usefulness of our approach, and shed light on how the compositional design provides an interpretable interface for more practical applications.
Abstract:Graph neural networks (GNNs) have achieved remarkable empirical success in processing and representing graph-structured data across various domains. However, a significant challenge known as "oversmoothing" persists, where vertex features become nearly indistinguishable in deep GNNs, severely restricting their expressive power and practical utility. In this work, we analyze the asymptotic oversmoothing rates of deep GNNs with and without residual connections by deriving explicit convergence rates for a normalized vertex similarity measure. Our analytical framework is grounded in the multiplicative ergodic theorem. Furthermore, we demonstrate that adding residual connections effectively mitigates or prevents oversmoothing across several broad families of parameter distributions. The theoretical findings are strongly supported by numerical experiments.
Abstract:The advancement of dense visual simultaneous localization and mapping (SLAM) has been greatly facilitated by the emergence of neural implicit representations. Neural implicit encoding SLAM, a typical example of which is NICE-SLAM, has recently demonstrated promising results in large-scale indoor scenes. However, these methods typically rely on temporally dense RGB-D image streams as input in order to function properly. When the input source does not support high frame rates or the camera movement is too fast, these methods often experience crashes or significant degradation in tracking and mapping accuracy. In this paper, we propose EvenNICER-SLAM, a novel approach that addresses this issue through the incorporation of event cameras. Event cameras are bio-inspired cameras that respond to intensity changes instead of absolute brightness. Specifically, we integrated an event loss backpropagation stream into the NICE-SLAM pipeline to enhance camera tracking with insufficient RGB-D input. We found through quantitative evaluation that EvenNICER-SLAM, with an inclusion of higher-frequency event image input, significantly outperforms NICE-SLAM with reduced RGB-D input frequency. Our results suggest the potential for event cameras to improve the robustness of dense SLAM systems against fast camera motion in real-world scenarios.
Abstract:Multi-modal Large Language Models (MLLMs) have demonstrated remarkable performance on various visual-language understanding and generation tasks. However, MLLMs occasionally generate content inconsistent with the given images, which is known as "hallucination". Prior works primarily center on evaluating hallucination using standard, unperturbed benchmarks, which overlook the prevalent occurrence of perturbed inputs in real-world scenarios-such as image cropping or blurring-that are critical for a comprehensive assessment of MLLMs' hallucination. In this paper, to bridge this gap, we propose Hallu-PI, the first benchmark designed to evaluate Hallucination in MLLMs within Perturbed Inputs. Specifically, Hallu-PI consists of seven perturbed scenarios, containing 1,260 perturbed images from 11 object types. Each image is accompanied by detailed annotations, which include fine-grained hallucination types, such as existence, attribute, and relation. We equip these annotations with a rich set of questions, making Hallu-PI suitable for both discriminative and generative tasks. Extensive experiments on 12 mainstream MLLMs, such as GPT-4V and Gemini-Pro Vision, demonstrate that these models exhibit significant hallucinations on Hallu-PI, which is not observed in unperturbed scenarios. Furthermore, our research reveals a severe bias in MLLMs' ability to handle different types of hallucinations. We also design two baselines specifically for perturbed scenarios, namely Perturbed-Reminder and Perturbed-ICL. We hope that our study will bring researchers' attention to the limitations of MLLMs when dealing with perturbed inputs, and spur further investigations to address this issue. Our code and datasets are publicly available at https://github.com/NJUNLP/Hallu-PI.
Abstract:3D human pose estimation is a vital task in computer vision, involving the prediction of human joint positions from images or videos to reconstruct a skeleton of a human in three-dimensional space. This technology is pivotal in various fields, including animation, security, human-computer interaction, and automotive safety, where it promotes both technological progress and enhanced human well-being. The advent of deep learning significantly advances the performance of 3D pose estimation by incorporating temporal information for predicting the spatial positions of human joints. However, traditional methods often fall short as they primarily focus on the spatial coordinates of joints and overlook the orientation and rotation of the connecting bones, which are crucial for a comprehensive understanding of human pose in 3D space. To address these limitations, we introduce Quater-GCN (Q-GCN), a directed graph convolutional network tailored to enhance pose estimation by orientation. Q-GCN excels by not only capturing the spatial dependencies among node joints through their coordinates but also integrating the dynamic context of bone rotations in 2D space. This approach enables a more sophisticated representation of human poses by also regressing the orientation of each bone in 3D space, moving beyond mere coordinate prediction. Furthermore, we complement our model with a semi-supervised training strategy that leverages unlabeled data, addressing the challenge of limited orientation ground truth data. Through comprehensive evaluations, Q-GCN has demonstrated outstanding performance against current state-of-the-art methods.
Abstract:The study of action recognition has attracted considerable attention recently due to its broad applications in multiple areas. However, with the issue of discontinuous training video, which not only decreases the performance of action recognition model, but complicates the data augmentation process as well, still remains under-exploration. In this study, we introduce the 4A (Action Animation-based Augmentation Approach), an innovative pipeline for data augmentation to address the problem. The main contributions remain in our work includes: (1) we investigate the problem of severe decrease on performance of action recognition task training by discontinuous video, and the limitation of existing augmentation methods on solving this problem. (2) we propose a novel augmentation pipeline, 4A, to address the problem of discontinuous video for training, while achieving a smoother and natural-looking action representation than the latest data augmentation methodology. (3) We achieve the same performance with only 10% of the original data for training as with all of the original data from the real-world dataset, and a better performance on In-the-wild videos, by employing our data augmentation techniques.
Abstract:Score-based Generative Models (SGMs) is one leading method in generative modeling, renowned for their ability to generate high-quality samples from complex, high-dimensional data distributions. The method enjoys empirical success and is supported by rigorous theoretical convergence properties. In particular, it has been shown that SGMs can generate samples from a distribution that is close to the ground-truth if the underlying score function is learned well, suggesting the success of SGM as a generative model. We provide a counter-example in this paper. Through the sample complexity argument, we provide one specific setting where the score function is learned well. Yet, SGMs in this setting can only output samples that are Gaussian blurring of training data points, mimicking the effects of kernel density estimation. The finding resonates a series of recent finding that reveal that SGMs can demonstrate strong memorization effect and fail to generate.
Abstract:Current datasets for action recognition tasks face limitations stemming from traditional collection and generation methods, including the constrained range of action classes, absence of multi-viewpoint recordings, limited diversity, poor video quality, and labor-intensive manually collection. To address these challenges, we introduce GTAutoAct, a innovative dataset generation framework leveraging game engine technology to facilitate advancements in action recognition. GTAutoAct excels in automatically creating large-scale, well-annotated datasets with extensive action classes and superior video quality. Our framework's distinctive contributions encompass: (1) it innovatively transforms readily available coordinate-based 3D human motion into rotation-orientated representation with enhanced suitability in multiple viewpoints; (2) it employs dynamic segmentation and interpolation of rotation sequences to create smooth and realistic animations of action; (3) it offers extensively customizable animation scenes; (4) it implements an autonomous video capture and processing pipeline, featuring a randomly navigating camera, with auto-trimming and labeling functionalities. Experimental results underscore the framework's robustness and highlights its potential to significantly improve action recognition model training.