Sorry, you need to enable JavaScript to visit this website.

Our objective is audio-visual synchronization with a focus on ‘in-the-wild’ videos, such as those on YouTube, where synchronization cues can be sparse. Our contributions include a novel audio-visual synchronization model, and training that decouples feature extraction from synchronization modelling through multi-modal segment-level contrastive pre-training. This approach achieves state-of-the-art performance in both dense and sparse settings.

Categories:
28 Views

Background music (BGM) can enhance the video’s emotion and thus make it engaging. However, selecting an appropriate BGM often requires domain knowledge or a deep understanding of the video. This has led to the development of video-music retrieval techniques. Most existing approaches utilize pre-trained video/music feature extractors trained with different target sets to obtain average video/music-level embeddings for cross-modal matching. The drawbacks are two-fold. One is that different target sets for video/music pre-training may cause the generated embeddings difficult to match.

Categories:
23 Views

Monocular 3D human pose estimation poses significant challenges due to the inherent depth ambiguities that arise during the reprojection process from 2D to 3D. Conventional approaches that rely on estimating an over-fit projection matrix struggle to effectively address these challenges and often result in noisy outputs. Recent advancements in diffusion models have shown promise in incorporating structural priors to address reprojection ambiguities.

Categories:
48 Views

Narrative understanding is an integrative task of studying characters, plots, events, and relations in a story.
It involves natural language processing tasks such as named entity recognition and coreference resolution to identify the characters, semantic role labeling and argument mining to find character actions and events, information extraction and question answering to describe character attributes, causal analysis to relate different events, and summarization to find the main storyline.

Categories:
48 Views

Visual Question Answering (VQA) is a task that requires models to comprehend both questions and images. An increasing number of works are leveraging the strong reasoning capabilities of Large Language Models (LLMs) to address VQA. These methods typically utilize image captions as visual text description to aid LLMs in comprehending images. However, these captions often overlooking the relations of fine-grained objects, which will limit the reasoning capability of LLMs. In this paper, we present PFVR, a modular framework that Prompts LLMs with Fine-grained Visual Relationships for VQA.

Categories:
54 Views

With the growing applications of water operations, water surface object detection tasks are facing new challenges. In this paper, we focus on improving the performance of water surface small object detection. Due to the limitations of single sensor in water environments, we propose RCFNet, a novel small object detection method based on radar-vision fusion. RCFNet fuses features captured by radar and camera in multiple stages to generate more effective target feature representations for small object detection on water surfaces.

Categories:
52 Views

Inspired by multi-task learning, degraded low-quality color-depth images enhancement tasks are transformed as a joint color-depth optimization model by using maximum a posteriori estimation. This model is optimized alternatively in an iterative way to get the solutions of CGD-SR task and Low-Brightness Color Image Enhancement (LBC-IE) task. The whole iterative optimization procedure is expanded as a joint model-driven unfolding network.

Categories:
28 Views

A new method of Gaussian process dynamic model (GPDM), named class-aware shared GPDM (CSGPDM), is presented in this paper. One of the most difference between our CSGPDM and existing GPDM is considering class information which helps to build the class label-based latent space being effective for the following class-related tasks. In terms of representation learning, CSGPDM is optimized by considering not only a non-linear relationship but also time-series relation and discriminative information of each class label.

Categories:
16 Views

Lack of audio-video synchronization is a common problem during television broadcasts and video conferencing, leading to an unsatisfactory viewing experience. A widely accepted paradigm is to create an error detection mechanism that identifies the cases when audio is leading or lagging. We propose ModEFormer, which independently extracts audio and video embeddings using modality-specific transformers.

Categories:
47 Views

Pages