Audio captioning is an important research area that aims to generate meaningful descriptions for audio clips. Most of the existing research extracts acoustic features of audio clips as input to encoder-decoder and transformer architectures to produce the captions in a sequence-to-sequence manner. Due to data insufficiency and the architecture's inadequate learning capacity, additional information is needed to generate natural language sentences, as well as acoustic features. To address these problems, an encoder-decoder architecture is proposed that learns from both acoustic features and extracted audio event labels as inputs. The proposed model is based on pre-trained acoustic features and audio event detection. Various experiments used different acoustic features, word embedding models, audio event label extraction methods, and implementation configurations to show which combinations have better performance on the audio captioning task. Results of the extensive experiments on multiple datasets show that using audio event labels with the acoustic features improves the recognition performance and the proposed method either outperforms or achieves competitive results with the state-of-the-art models.
相關內容
Graph convolutional networks have made great progress in graph-based semi-supervised learning. Existing methods mainly assume that nodes connected by graph edges are prone to have similar attributes and labels, so that the features smoothed by local graph structures can reveal the class similarities. However, there often exist mismatches between graph structures and labels in many real-world scenarios, where the structures may propagate misleading features or labels that eventually affect the model performance. In this paper, we propose a multi-task self-distillation framework that injects self-supervised learning and self-distillation into graph convolutional networks to separately address the mismatch problem from the structure side and the label side. First, we formulate a self-supervision pipeline based on pre-text tasks to capture different levels of similarities in graphs. The feature extraction process is encouraged to capture more complex proximity by jointly optimizing the pre-text task and the target task. Consequently, the local feature aggregations are improved from the structure side. Second, self-distillation uses soft labels of the model itself as additional supervision, which has similar effects as label smoothing. The knowledge from the classification pipeline and the self-supervision pipeline is collectively distilled to improve the generalization ability of the model from the label side. Experiment results show that the proposed method obtains remarkable performance gains under several classic graph convolutional architectures.
Creation of images using generative adversarial networks has been widely adapted into multi-modal regime with the advent of multi-modal representation models pre-trained on large corpus. Various modalities sharing a common representation space could be utilized to guide the generative models to create images from text or even from audio source. Departing from the previous methods that solely rely on either text or audio, we exploit the expressiveness of both modality. Based on the fusion of text and audio, we create video whose content is consistent with the distinct modalities that are provided. A simple approach to automatically segment the video into variable length intervals and maintain time consistency in generated video is part of our method. Our proposed framework for generating music video shows promising results in application level where users can interactively feed in music source and text source to create artistic music videos. Our code is available at //github.com/joeljang/music2video.
Connecting Vision and Language plays an essential role in Generative Intelligence. For this reason, in the last few years, a large research effort has been devoted to image captioning, i.e. the task of describing images with syntactically and semantically meaningful sentences. Starting from 2015 the task has generally been addressed with pipelines composed of a visual encoding step and a language model for text generation. During these years, both components have evolved considerably through the exploitation of object regions, attributes, and relationships and the introduction of multi-modal connections, fully-attentive approaches, and BERT-like early-fusion strategies. However, regardless of the impressive results obtained, research in image captioning has not reached a conclusive answer yet. This work aims at providing a comprehensive overview and categorization of image captioning approaches, from visual encoding and text generation to training strategies, used datasets, and evaluation metrics. In this respect, we quantitatively compare many relevant state-of-the-art approaches to identify the most impactful technical innovations in image captioning architectures and training strategies. Moreover, many variants of the problem and its open challenges are analyzed and discussed. The final goal of this work is to serve as a tool for understanding the existing state-of-the-art and highlighting the future directions for an area of research where Computer Vision and Natural Language Processing can find an optimal synergy.
Video captioning is a challenging task that requires a deep understanding of visual scenes. State-of-the-art methods generate captions using either scene-level or object-level information but without explicitly modeling object interactions. Thus, they often fail to make visually grounded predictions, and are sensitive to spurious correlations. In this paper, we propose a novel spatio-temporal graph model for video captioning that exploits object interactions in space and time. Our model builds interpretable links and is able to provide explicit visual grounding. To avoid unstable performance caused by the variable number of objects, we further propose an object-aware knowledge distillation mechanism, in which local object information is used to regularize global scene features. We demonstrate the efficacy of our approach through extensive experiments on two benchmarks, showing our approach yields competitive performance with interpretable predictions.
UniViLM: A Unified Video and Language Pre-Training Model for Multimodal Understanding and Generation
We propose UniViLM: a Unified Video and Language pre-training Model for multimodal understanding and generation. Motivated by the recent success of BERT based pre-training technique for NLP and image-language tasks, VideoBERT and CBT are proposed to exploit BERT model for video and language pre-training using narrated instructional videos. Different from their works which only pre-train understanding task, we propose a unified video-language pre-training model for both understanding and generation tasks. Our model comprises of 4 components including two single-modal encoders, a cross encoder and a decoder with the Transformer backbone. We first pre-train our model to learn the universal representation for both video and language on a large instructional video dataset. Then we fine-tune the model on two multimodal tasks including understanding task (text-based video retrieval) and generation task (multimodal video captioning). Our extensive experiments show that our method can improve the performance of both understanding and generation tasks and achieves the state-of-the art results.
Textual entailment is a fundamental task in natural language processing. Most approaches for solving the problem use only the textual content present in training data. A few approaches have shown that information from external knowledge sources like knowledge graphs (KGs) can add value, in addition to the textual content, by providing background knowledge that may be critical for a task. However, the proposed models do not fully exploit the information in the usually large and noisy KGs, and it is not clear how it can be effectively encoded to be useful for entailment. We present an approach that complements text-based entailment models with information from KGs by (1) using Personalized PageR- ank to generate contextual subgraphs with reduced noise and (2) encoding these subgraphs using graph convolutional networks to capture KG structure. Our technique extends the capability of text models exploiting structural and semantic information found in KGs. We evaluate our approach on multiple textual entailment datasets and show that the use of external knowledge helps improve prediction accuracy. This is particularly evident in the challenging BreakingNLI dataset, where we see an absolute improvement of 5-20% over multiple text-based entailment models.
We propose a new method for event extraction (EE) task based on an imitation learning framework, specifically, inverse reinforcement learning (IRL) via generative adversarial network (GAN). The GAN estimates proper rewards according to the difference between the actions committed by the expert (or ground truth) and the agent among complicated states in the environment. EE task benefits from these dynamic rewards because instances and labels yield to various extents of difficulty and the gains are expected to be diverse -- e.g., an ambiguous but correctly detected trigger or argument should receive high gains -- while the traditional RL models usually neglect such differences and pay equal attention on all instances. Moreover, our experiments also demonstrate that the proposed framework outperforms state-of-the-art methods, without explicit feature engineering.
Dense video captioning aims to generate text descriptions for all events in an untrimmed video. This involves both detecting and describing events. Therefore, all previous methods on dense video captioning tackle this problem by building two models, i.e. an event proposal and a captioning model, for these two sub-problems. The models are either trained separately or in alternation. This prevents direct influence of the language description to the event proposal, which is important for generating accurate descriptions. To address this problem, we propose an end-to-end transformer model for dense video captioning. The encoder encodes the video into appropriate representations. The proposal decoder decodes from the encoding with different anchors to form video event proposals. The captioning decoder employs a masking network to restrict its attention to the proposal event over the encoding feature. This masking network converts the event proposal to a differentiable mask, which ensures the consistency between the proposal and captioning during training. In addition, our model employs a self-attention mechanism, which enables the use of efficient non-recurrent structure during encoding and leads to performance improvements. We demonstrate the effectiveness of this end-to-end model on ActivityNet Captions and YouCookII datasets, where we achieved 10.12 and 6.58 METEOR score, respectively.
Automatic image captioning has recently approached human-level performance due to the latest advances in computer vision and natural language understanding. However, most of the current models can only generate plain factual descriptions about the content of a given image. However, for human beings, image caption writing is quite flexible and diverse, where additional language dimensions, such as emotion, humor and language styles, are often incorporated to produce diverse, emotional, or appealing captions. In particular, we are interested in generating sentiment-conveying image descriptions, which has received little attention. The main challenge is how to effectively inject sentiments into the generated captions without altering the semantic matching between the visual content and the generated descriptions. In this work, we propose two different models, which employ different schemes for injecting sentiments into image captions. Compared with the few existing approaches, the proposed models are much simpler and yet more effective. The experimental results show that our model outperform the state-of-the-art models in generating sentimental (i.e., sentiment-bearing) image captions. In addition, we can also easily manipulate the model by assigning different sentiments to the testing image to generate captions with the corresponding sentiments.
Most previous event extraction studies have relied heavily on features derived from annotated event mentions, thus cannot be applied to new event types without annotation effort. In this work, we take a fresh look at event extraction and model it as a grounding problem. We design a transferable neural architecture, mapping event mentions and types jointly into a shared semantic space using structural and compositional neural networks, where the type of each event mention can be determined by the closest of all candidate types . By leveraging (1)~available manual annotations for a small set of existing event types and (2)~existing event ontologies, our framework applies to new event types without requiring additional annotation. Experiments on both existing event types (e.g., ACE, ERE) and new event types (e.g., FrameNet) demonstrate the effectiveness of our approach. \textit{Without any manual annotations} for 23 new event types, our zero-shot framework achieved performance comparable to a state-of-the-art supervised model which is trained from the annotations of 500 event mentions.
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