Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userXGPT: Cross-modal Generative Pre-Training for Image Captioning
95
0
0.0
(
0
)
Added by
Qiaolin Xia
Publication date
2020
fields
Informatics Engineering
and research's language is
English
Ask ChatGPT about the research
No Arabic abstract
While many BERT-based cross-modal pre-trained models produce excellent results on downstream understanding tasks like image-text retrieval and VQA, they cannot be applied to generation tasks directly. In this paper, we propose XGPT, a new method of Cross-modal Generative Pre-Training for Image Captioning that is designed to pre-train text-to-image caption generators through three novel generation tasks, including Image-conditioned Masked Language Modeling (IMLM), Image-conditioned Denoising Autoencoding (IDA), and Text-conditioned Image Feature Generation (TIFG). As a result, the pre-trained XGPT can be fine-tuned without any task-specific architecture modifications to create state-of-the-art models for image captioning. Experiments show that XGPT obtains new state-of-the-art results on the benchmark datasets, including COCO Captions and Flickr30k Captions. We also use XGPT to generate new image captions as data augmentation for the image retrieval task and achieve significant improvement on all recall metrics.
rate research
Read More
While large scale pre-training has achieved great achievements in bridging the gap between vision and language, it still faces several challenges. First, the cost for pre-training is expensive. Second, there is no efficient way to handle the data noise which degrades model performance. Third, previous methods only leverage limited image-text paired data, while ignoring richer single-modal data, which may result in poor generalization to single-modal downstream tasks. In this work, we propose an EfficientCLIP method via Ensemble Confident Learning to obtain a less noisy data subset. Extra rich non-paired single-modal text data is used for boosting the generalization of text branch. We achieve the state-of-the-art performance on Chinese cross-modal retrieval tasks with only 1/10 training resources compared to CLIP and WenLan, while showing excellent generalization to single-modal tasks, including text retrieval and text classification.
In recent years, the biggest advances in major Computer Vision tasks, such as object recognition, handwritten-digit identification, facial recognition, and many others., have all come through the use of Convolutional Neural Networks (CNNs). Similarly, in the domain of Natural Language Processing, Recurrent Neural Networks (RNNs), and Long Short Term Memory networks (LSTMs) in particular, have been crucial to some of the biggest breakthroughs in performance for tasks such as machine translation, part-of-speech tagging, sentiment analysis, and many others. These individual advances have greatly benefited tasks even at the intersection of NLP and Computer Vision, and inspired by this success, we studied some existing neural image captioning models that have proven to work well. In this work, we study some existing captioning models that provide near state-of-the-art performances, and try to enhance one such model. We also present a simple image captioning model that makes use of a CNN, an LSTM, and the beam search1 algorithm, and study its performance based on various qualitative and quantitative metrics.
Research in image captioning has mostly focused on English because of the availability of image-caption paired datasets in this language. However, building vision-language systems only for English deprives a large part of the world population of AI technologies benefit. On the other hand, creating image-caption paired datasets for every target language is expensive. In this work, we present a novel unsupervised cross-lingual method to generate image captions in a target language without using any image-caption corpus in the source or target languages. Our method relies on (i) a cross-lingual scene graph to sentence translation process, which learns to decode sentences in the target language from a cross-lingual encoding space of scene graphs using a sentence parallel (bitext) corpus, and (ii) an unsupervised cross-modal feature mapping which seeks to map an encoded scene graph features from image modality to language modality. We verify the effectiveness of our proposed method on the Chinese image caption generation task. The comparisons against several existing methods demonstrate the effectiveness of our approach.
Multimodal pre-training has propelled great advancement in vision-and-language research. These large-scale pre-trained models, although successful, fatefully suffer from slow inference speed due to enormous computation cost mainly from cross-modal attention in Transformer architecture. When applied to real-life applications, such latency and computation demand severely deter the practical use of pre-trained models. In this paper, we study Image-text retrieval (ITR), the most mature scenario of V+L application, which has been widely studied even prior to the emergence of recent pre-trained models. We propose a simple yet highly effective approach, LightningDOT that accelerates the inference time of ITR by thousands of times, without sacrificing accuracy. LightningDOT removes the time-consuming cross-modal attention by pre-training on three novel learning objectives, extracting feature indexes offline, and employing instant dot-product matching with further re-ranking, which significantly speeds up retrieval process. In fact, LightningDOT achieves new state of the art across multiple ITR benchmarks such as Flickr30k, COCO and Multi30K, outperforming existing pre-trained models that consume 1000x magnitude of computational hours. Code and pre-training checkpoints are available at https://github.com/intersun/LightningDOT.
Language model pre-training has shown promising results in various downstream tasks. In this context, we introduce a cross-modal pre-trained language model, called Speech-Text BERT (ST-BERT), to tackle end-to-end spoken language understanding (E2E SLU) tasks. Taking phoneme posterior and subword-level text as an input, ST-BERT learns a contextualized cross-modal alignment via our two proposed pre-training tasks: Cross-modal Masked Language Modeling (CM-MLM) and Cross-modal Conditioned Language Modeling (CM-CLM). Experimental results on three benchmarks present that our approach is effective for various SLU datasets and shows a surprisingly marginal performance degradation even when 1% of the training data are available. Also, our method shows further SLU performance gain via domain-adaptive pre-training with domain-specific speech-text pair data.
suggested questions
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
