Quality assessment of images and videos emphasizes both local details and global semantics, whereas general data sampling methods (e.g., resizing, cropping or grid-based fragment) fail to catch them simultaneously. To address the deficiency, current approaches have to adopt multi-branch models and take as input the multi-resolution data, which burdens the model complexity. In this work, instead of stacking up models, a more elegant data sampling method (named as SAMA, scaling and masking) is explored, which compacts both the local and global content in a regular input size. The basic idea is to scale the data into a pyramid first, and reduce the pyramid into a regular data dimension with a masking strategy. Benefiting from the spatial and temporal redundancy in images and videos, the processed data maintains the multi-scale characteristics with a regular input size, thus can be processed by a single-branch model. We verify the sampling method in image and video quality assessment. Experiments show that our sampling method can improve the performance of current single-branch models significantly, and achieves competitive performance to the multi-branch models without extra model complexity. The source code will be available at //github.com/Sissuire/SAMA.
相關內容
Most video captioning models are designed to process short video clips of few seconds and output text describing low-level visual concepts (e.g., objects, scenes, atomic actions). However, most real-world videos last for minutes or hours and have a complex hierarchical structure spanning different temporal granularities. We propose Video ReCap, a recursive video captioning model that can process video inputs of dramatically different lengths (from 1 second to 2 hours) and output video captions at multiple hierarchy levels. The recursive video-language architecture exploits the synergy between different video hierarchies and can process hour-long videos efficiently. We utilize a curriculum learning training scheme to learn the hierarchical structure of videos, starting from clip-level captions describing atomic actions, then focusing on segment-level descriptions, and concluding with generating summaries for hour-long videos. Furthermore, we introduce Ego4D-HCap dataset by augmenting Ego4D with 8,267 manually collected long-range video summaries. Our recursive model can flexibly generate captions at different hierarchy levels while also being useful for other complex video understanding tasks, such as VideoQA on EgoSchema. Data, code, and models are available at: //sites.google.com/view/vidrecap
Natural language and search interfaces intuitively facilitate data exploration and provide visualization responses to diverse analytical queries based on the underlying datasets. However, these interfaces often fail to interpret more complex analytical intents, such as discerning subtleties and quantifiable differences between terms like "bump" and "spike" in the context of COVID cases, for example. We address this gap by extending the capabilities of a data exploration search interface for interpreting semantic concepts in time series trends. We first create a comprehensive dataset of semantic concepts by mapping quantifiable univariate data trends such as slope and angle to crowdsourced, semantically meaningful trend labels. The dataset contains quantifiable properties that capture the slope-scalar effect of semantic modifiers like "sharply" and "gradually," as well as multi-line trends (e.g., "peak," "valley"). We demonstrate the utility of this dataset in SlopeSeeker, a tool that supports natural language querying of quantifiable trends, such as "show me stocks that tanked in 2010." The tool incorporates novel scoring and ranking techniques based on semantic relevance and visual prominence to present relevant trend chart responses containing these semantic trend concepts. In addition, SlopeSeeker provides a faceted search interface for users to navigate a semantic hierarchy of concepts from general trends (e.g., "increase") to more specific ones (e.g., "sharp increase"). A preliminary user evaluation of the tool demonstrates that the search interface supports greater expressivity of queries containing concepts that describe data trends. We identify potential future directions for leveraging our publicly available quantitative semantics dataset in other data domains and for novel visual analytics interfaces.
We analyze the behaviors of open large language models (LLMs) on the task of data-to-text (D2T) generation, i.e., generating coherent and relevant text from structured data. To avoid the issue of LLM training data contamination with standard benchmarks, we design Quintd - a tool for collecting novel structured data records from public APIs. Using a dataset collected with Quintd and leveraging reference-free evaluation, we analyze model behaviors on five D2T generation tasks. We find that recent open LLMs (Llama2, Mistral, and Zephyr) can generate fluent and coherent text from standard data formats in zero-shot settings. However, we also show that the semantic accuracy of the outputs is a major issue: both according to our GPT-4-based metric and human annotators, more than 80% of the outputs of open LLMs contain a semantic error. We publicly release the code, data, and model outputs.
The integration of large language models (LLMs) and search engines represents a significant evolution in knowledge acquisition methodologies. However, determining the knowledge that an LLM already possesses and the knowledge that requires the help of a search engine remains an unresolved issue. Most existing methods solve this problem through the results of preliminary answers or reasoning done by the LLM itself, but this incurs excessively high computational costs. This paper introduces a novel collaborative approach, namely SlimPLM, that detects missing knowledge in LLMs with a slim proxy model, to enhance the LLM's knowledge acquisition process. We employ a proxy model which has far fewer parameters, and take its answers as heuristic answers. Heuristic answers are then utilized to predict the knowledge required to answer the user question, as well as the known and unknown knowledge within the LLM. We only conduct retrieval for the missing knowledge in questions that the LLM does not know. Extensive experimental results on five datasets with two LLMs demonstrate a notable improvement in the end-to-end performance of LLMs in question-answering tasks, achieving or surpassing current state-of-the-art models with lower LLM inference costs.
The creation of accurate virtual models of real-world objects is imperative to robotic simulations and applications such as computer vision, artificial intelligence, and machine learning. This paper documents the different methods employed for generating a database of mesh models of real-world objects. These methods address the tedious and time-intensive process of manually generating the models using CAD software. Essentially, DSLR/phone cameras were employed to acquire images of target objects. These images were processed using a photogrammetry software known as Meshroom to generate a dense surface reconstruction of the scene. The result produced by Meshroom was edited and simplified using MeshLab, a mesh-editing software to produce the final model. Based on the obtained models, this process was effective in modelling the geometry and texture of real-world objects with high fidelity. An active 3D scanner was also utilized to accelerate the process for large objects. All generated models and captured images are made available on the website of the project.
The text-driven image and video diffusion models have achieved unprecedented success in generating realistic and diverse content. Recently, the editing and variation of existing images and videos in diffusion-based generative models have garnered significant attention. However, previous works are limited to editing content with text or providing coarse personalization using a single visual clue, rendering them unsuitable for indescribable content that requires fine-grained and detailed control. In this regard, we propose a generic video editing framework called Make-A-Protagonist, which utilizes textual and visual clues to edit videos with the goal of empowering individuals to become the protagonists. Specifically, we leverage multiple experts to parse source video, target visual and textual clues, and propose a visual-textual-based video generation model that employs mask-guided denoising sampling to generate the desired output. Extensive results demonstrate the versatile and remarkable editing capabilities of Make-A-Protagonist.
Inpainting involves filling in missing pixels or areas in an image, a crucial technique employed in Mixed Reality environments for various applications, particularly in Diminished Reality (DR) where content is removed from a user's visual environment. Existing methods rely on digital replacement techniques which necessitate multiple cameras and incur high costs. AR devices and smartphones use ToF depth sensors to capture scene depth maps aligned with RGB images. Despite speed and affordability, ToF cameras create imperfect depth maps with missing pixels. To address the above challenges, we propose Hierarchical Inpainting GAN (HI-GAN), a novel approach comprising three GANs in a hierarchical fashion for RGBD inpainting. EdgeGAN and LabelGAN inpaint masked edge and segmentation label images respectively, while CombinedRGBD-GAN combines their latent representation outputs and performs RGB and Depth inpainting. Edge images and particularly segmentation label images as auxiliary inputs significantly enhance inpainting performance by complementary context and hierarchical optimization. We believe we make the first attempt to incorporate label images into inpainting process.Unlike previous approaches requiring multiple sequential models and separate outputs, our work operates in an end-to-end manner, training all three models simultaneously and hierarchically. Specifically, EdgeGAN and LabelGAN are first optimized separately and further optimized inside CombinedRGBD-GAN to enhance inpainting quality. Experiments demonstrate that HI-GAN works seamlessly and achieves overall superior performance compared with existing approaches.
Knowledge graph reasoning (KGR), aiming to deduce new facts from existing facts based on mined logic rules underlying knowledge graphs (KGs), has become a fast-growing research direction. It has been proven to significantly benefit the usage of KGs in many AI applications, such as question answering and recommendation systems, etc. According to the graph types, the existing KGR models can be roughly divided into three categories, \textit{i.e.,} static models, temporal models, and multi-modal models. The early works in this domain mainly focus on static KGR and tend to directly apply general knowledge graph embedding models to the reasoning task. However, these models are not suitable for more complex but practical tasks, such as inductive static KGR, temporal KGR, and multi-modal KGR. To this end, multiple works have been developed recently, but no survey papers and open-source repositories comprehensively summarize and discuss models in this important direction. To fill the gap, we conduct a survey for knowledge graph reasoning tracing from static to temporal and then to multi-modal KGs. Concretely, the preliminaries, summaries of KGR models, and typical datasets are introduced and discussed consequently. Moreover, we discuss the challenges and potential opportunities. The corresponding open-source repository is shared on GitHub: //github.com/LIANGKE23/Awesome-Knowledge-Graph-Reasoning.
Graph neural networks generalize conventional neural networks to graph-structured data and have received widespread attention due to their impressive representation ability. In spite of the remarkable achievements, the performance of Euclidean models in graph-related learning is still bounded and limited by the representation ability of Euclidean geometry, especially for datasets with highly non-Euclidean latent anatomy. Recently, hyperbolic space has gained increasing popularity in processing graph data with tree-like structure and power-law distribution, owing to its exponential growth property. In this survey, we comprehensively revisit the technical details of the current hyperbolic graph neural networks, unifying them into a general framework and summarizing the variants of each component. More importantly, we present various HGNN-related applications. Last, we also identify several challenges, which potentially serve as guidelines for further flourishing the achievements of graph learning in hyperbolic spaces.
In order to answer natural language questions over knowledge graphs, most processing pipelines involve entity and relation linking. Traditionally, entity linking and relation linking has been performed either as dependent sequential tasks or independent parallel tasks. In this paper, we propose a framework called "EARL", which performs entity linking and relation linking as a joint single task. EARL uses a graph connection based solution to the problem. We model the linking task as an instance of the Generalised Travelling Salesman Problem (GTSP) and use GTSP approximate algorithm solutions. We later develop EARL which uses a pair-wise graph-distance based solution to the problem.The system determines the best semantic connection between all keywords of the question by referring to a knowledge graph. This is achieved by exploiting the "connection density" between entity candidates and relation candidates. The "connection density" based solution performs at par with the approximate GTSP solution.We have empirically evaluated the framework on a dataset with 5000 questions. Our system surpasses state-of-the-art scores for entity linking task by reporting an accuracy of 0.65 to 0.40 from the next best entity linker.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191