In existing Video Frame Interpolation (VFI) approaches, the motion estimation between neighboring frames plays a crucial role. However, the estimation accuracy in existing methods remains a challenge, primarily due to the inherent ambiguity in identifying corresponding areas in adjacent frames for interpolation. Therefore, enhancing accuracy by distinguishing different regions before motion estimation is of utmost importance. In this paper, we introduce a novel solution involving the utilization of open-world segmentation models, e.g., SAM (Segment Anything Model), to derive Region-Distinguishable Priors (RDPs) in different frames. These RDPs are represented as spatial-varying Gaussian mixtures, distinguishing an arbitrary number of areas with a unified modality. RDPs can be integrated into existing motion-based VFI methods to enhance features for motion estimation, facilitated by our designed play-and-plug Hierarchical Region-aware Feature Fusion Module (HRFFM). HRFFM incorporates RDP into various hierarchical stages of VFI's encoder, using RDP-guided Feature Normalization (RDPFN) in a residual learning manner. With HRFFM and RDP, the features within VFI's encoder exhibit similar representations for matched regions in neighboring frames, thus improving the synthesis of intermediate frames. Extensive experiments demonstrate that HRFFM consistently enhances VFI performance across various scenes.
相關內容
Learning from Demonstration (LfD) is a widely used technique for skill acquisition in robotics. However, demonstrations of the same skill may exhibit significant variances, or learning systems may attempt to acquire different means of the same skill simultaneously, making it challenging to encode these motions into movement primitives. To address these challenges, we propose an LfD framework, namely the Conditional Neural Expert Processes (CNEP), that learns to assign demonstrations from different modes to distinct expert networks utilizing the inherent information within the latent space to match experts with the encoded representations. CNEP does not require supervision on which mode the trajectories belong to. Provided experiments on artificially generated datasets demonstrate the efficacy of CNEP. Furthermore, we compare the performance of CNEP with another LfD framework, namely Conditional Neural Movement Primitives (CNMP), on a range of tasks, including experiments on a real robot. The results reveal enhanced modeling performance for movement primitives, leading to the synthesis of trajectories that more accurately reflect those demonstrated by experts, particularly when the model inputs include intersection points from various trajectories. Additionally, CNEP offers improved interpretability and faster convergence by promoting expert specialization. Furthermore, we show that the CNEP model accomplishes obstacle avoidance tasks with a real manipulator when provided with novel start and destination points, in contrast to the CNMP model, which leads to collisions with the obstacle.
Existing methods for controlling language models, such as RLHF and Constitutional AI, involve determining which LLM behaviors are desirable and training them into a language model. However, in many cases, it is desirable for LLMs to be controllable \textit{at inference time}, so that they can be used in multiple contexts with diverse needs. We illustrate this with the \textbf{Pink Elephant Problem}: instructing an LLM to avoid discussing a certain entity (a ``Pink Elephant''), and instead discuss a preferred entity (``Grey Elephant''). We apply a novel simplification of Constitutional AI, \textbf{Direct Principle Feedback}, which skips the ranking of responses and uses DPO directly on critiques and revisions. Our results show that after DPF fine-tuning on our synthetic Pink Elephants dataset, our 13B fine-tuned LLaMA 2 model significantly outperforms Llama-2-13B-Chat and a prompted baseline, and performs as well as GPT-4 in on our curated test set assessing the Pink Elephant Problem.
Human Pose Estimation (HPE) to assess human motion in sports, rehabilitation or work safety requires accurate sensing without compromising the sensitive underlying personal data. Therefore, local processing is necessary and the limited energy budget in such systems can be addressed by Inertial Measurement Units (IMU) instead of common camera sensing. The central trade-off between accuracy and efficient use of hardware resources is rarely discussed in research. We address this trade-off by a simulative Design Space Exploration (DSE) of a varying quantity and positioning of IMU-sensors. First, we generate IMU-data from a publicly available body model dataset for different sensor configurations and train a deep learning model with this data. Additionally, we propose a combined metric to assess the accuracy-resource trade-off. We used the DSE as a tool to evaluate sensor configurations and identify beneficial ones for a specific use case. Exemplary, for a system with equal importance of accuracy and resources, we identify an optimal sensor configuration of 4 sensors with a mesh error of 6.03 cm, increasing the accuracy by 32.7% and reducing the hardware effort by two sensors compared to state of the art. Our work can be used to design health applications with well-suited sensor positioning and attention to data privacy and resource-awareness.
Vision--Language Models (VLMs) have emerged as the dominant approach for zero-shot recognition, adept at handling diverse scenarios and significant distribution changes. However, their deployment in risk-sensitive areas requires a deeper understanding of their uncertainty estimation capabilities, a relatively uncharted area. In this study, we explore the calibration properties of VLMs across different architectures, datasets, and training strategies. In particular, we analyze the uncertainty estimation performance of VLMs when calibrated in one domain, label set or hierarchy level, and tested in a different one. Our findings reveal that while VLMs are not inherently calibrated for uncertainty, temperature scaling significantly and consistently improves calibration, even across shifts in distribution and changes in label set. Moreover, VLMs can be calibrated with a very small set of examples. Through detailed experimentation, we highlight the potential applications and importance of our insights, aiming for more reliable and effective use of VLMs in critical, real-world scenarios.
Addressing runtime uncertainties in Machine Learning-Enabled Systems (MLS) is crucial for maintaining Quality of Service (QoS). The Machine Learning Model Balancer is a concept that addresses these uncertainties by facilitating dynamic ML model switching, showing promise in improving QoS in MLS. Leveraging this concept, this paper introduces SWITCH, an exemplar developed to enhance self-adaptive capabilities in such systems through dynamic model switching in runtime. SWITCH is designed as a comprehensive web service catering to a broad range of ML scenarios, with its implementation demonstrated through an object detection use case. SWITCH provides researchers with a flexible platform to apply and evaluate their ML model switching strategies, aiming to enhance QoS in MLS. SWITCH features advanced input handling, real-time data processing, and logging for adaptation metrics supplemented with an interactive real-time dashboard for enhancing system observability. This paper details SWITCH's architecture, self-adaptation strategies through ML model switching, and its empirical validation through a case study, illustrating its potential to improve QoS in MLS. By enabling a hands-on approach to explore adaptive behaviors in ML systems, SWITCH contributes a valuable tool to the SEAMS community for research into self-adaptive mechanisms for MLS and their practical applications.
Creating digital avatars from textual prompts has long been a desirable yet challenging task. Despite the promising outcomes obtained through 2D diffusion priors in recent works, current methods face challenges in achieving high-quality and animated avatars effectively. In this paper, we present $\textbf{HeadStudio}$, a novel framework that utilizes 3D Gaussian splatting to generate realistic and animated avatars from text prompts. Our method drives 3D Gaussians semantically to create a flexible and achievable appearance through the intermediate FLAME representation. Specifically, we incorporate the FLAME into both 3D representation and score distillation: 1) FLAME-based 3D Gaussian splatting, driving 3D Gaussian points by rigging each point to a FLAME mesh. 2) FLAME-based score distillation sampling, utilizing FLAME-based fine-grained control signal to guide score distillation from the text prompt. Extensive experiments demonstrate the efficacy of HeadStudio in generating animatable avatars from textual prompts, exhibiting visually appealing appearances. The avatars are capable of rendering high-quality real-time ($\geq 40$ fps) novel views at a resolution of 1024. They can be smoothly controlled by real-world speech and video. We hope that HeadStudio can advance digital avatar creation and that the present method can widely be applied across various domains.
This paper tackles the problem of motion deblurring of dynamic scenes. Although end-to-end fully convolutional designs have recently advanced the state-of-the-art in non-uniform motion deblurring, their performance-complexity trade-off is still sub-optimal. Most existing approaches achieve a large receptive field by increasing the number of generic convolution layers and kernel size. In this work, we propose a pixel adaptive and feature attentive design for handling large blur variations across different spatial locations and process each test image adaptively. We design a content-aware global-local filtering module that significantly improves performance by considering not only global dependencies but also by dynamically exploiting neighboring pixel information. We further introduce a pixel-adaptive non-uniform sampling strategy that implicitly discovers the difficult-to-restore regions present in the image and, in turn, performs fine-grained refinement in a progressive manner. Extensive qualitative and quantitative comparisons with prior art on deblurring benchmarks demonstrate that our approach performs favorably against the state-of-the-art deblurring algorithms.
Graph Neural Networks (GNNs) are state-of-the-art models for performing prediction tasks on graphs. While existing GNNs have shown great performance on various tasks related to graphs, little attention has been paid to the scenario where out-of-distribution (OOD) nodes exist in the graph during training and inference. Borrowing the concept from CV and NLP, we define OOD nodes as nodes with labels unseen from the training set. Since a lot of networks are automatically constructed by programs, real-world graphs are often noisy and may contain nodes from unknown distributions. In this work, we define the problem of graph learning with out-of-distribution nodes. Specifically, we aim to accomplish two tasks: 1) detect nodes which do not belong to the known distribution and 2) classify the remaining nodes to be one of the known classes. We demonstrate that the connection patterns in graphs are informative for outlier detection, and propose Out-of-Distribution Graph Attention Network (OODGAT), a novel GNN model which explicitly models the interaction between different kinds of nodes and separate inliers from outliers during feature propagation. Extensive experiments show that OODGAT outperforms existing outlier detection methods by a large margin, while being better or comparable in terms of in-distribution classification.
We investigate the problem of automatically determining what type of shoe left an impression found at a crime scene. This recognition problem is made difficult by the variability in types of crime scene evidence (ranging from traces of dust or oil on hard surfaces to impressions made in soil) and the lack of comprehensive databases of shoe outsole tread patterns. We find that mid-level features extracted by pre-trained convolutional neural nets are surprisingly effective descriptors for this specialized domains. However, the choice of similarity measure for matching exemplars to a query image is essential to good performance. For matching multi-channel deep features, we propose the use of multi-channel normalized cross-correlation and analyze its effectiveness. Our proposed metric significantly improves performance in matching crime scene shoeprints to laboratory test impressions. We also show its effectiveness in other cross-domain image retrieval problems: matching facade images to segmentation labels and aerial photos to map images. Finally, we introduce a discriminatively trained variant and fine-tune our system through our proposed metric, obtaining state-of-the-art performance.
Visual Question Answering (VQA) models have struggled with counting objects in natural images so far. We identify a fundamental problem due to soft attention in these models as a cause. To circumvent this problem, we propose a neural network component that allows robust counting from object proposals. Experiments on a toy task show the effectiveness of this component and we obtain state-of-the-art accuracy on the number category of the VQA v2 dataset without negatively affecting other categories, even outperforming ensemble models with our single model. On a difficult balanced pair metric, the component gives a substantial improvement in counting over a strong baseline by 6.6%.
小貼士
登錄享
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191