In this report, we introduce our adaptation of image-text models for long-term action anticipation. Our Video + CLIP framework makes use of a large-scale pre-trained paired image-text model: CLIP and a video encoder Slowfast network. The CLIP embedding provides fine-grained understanding of objects relevant for an action whereas the slowfast network is responsible for modeling temporal information within a video clip of few frames. We show that the features obtained from both encoders are complementary to each other, thus outperforming the baseline on Ego4D for the task of long-term action anticipation. Our code is available at github.com/srijandas07/clip_baseline_LTA_Ego4d.
Existing methods for arbitrary-shaped text detection in natural scenes face two critical issues, i.e., 1) fracture detections at the gaps in a text instance; and 2) inaccurate detections of arbitrary-shaped text instances with diverse background context. To address these issues, we propose a novel method named Intra- and Inter-Instance Collaborative Learning (I3CL). Specifically, to address the first issue, we design an effective convolutional module with multiple receptive fields, which is able to collaboratively learn better character and gap feature representations at local and long ranges inside a text instance. To address the second issue, we devise an instance-based transformer module to exploit the dependencies between different text instances and a global context module to exploit the semantic context from the shared background, which are able to collaboratively learn more discriminative text feature representation. In this way, I3CL can effectively exploit the intra- and inter-instance dependencies together in a unified end-to-end trainable framework. Besides, to make full use of the unlabeled data, we design an effective semi-supervised learning method to leverage the pseudo labels via an ensemble strategy. Without bells and whistles, experimental results show that the proposed I3CL sets new state-of-the-art results on three challenging public benchmarks, i.e., an F-measure of 77.5% on ICDAR2019-ArT, 86.9% on Total-Text, and 86.4% on CTW-1500. Notably, our I3CL with the ResNeSt-101 backbone ranked 1st place on the ICDAR2019-ArT leaderboard. The source code will be made publicly available.
Data augmentation aims to enrich training samples for alleviating the overfitting issue in low-resource or class-imbalanced situations. Traditional methods first devise task-specific operations such as Synonym Substitute, then preset the corresponding parameters such as the substitution rate artificially, which require a lot of prior knowledge and are prone to fall into the sub-optimum. Besides, the number of editing operations is limited in the previous methods, which decreases the diversity of the augmented data and thus restricts the performance gain. To overcome the above limitations, we propose a framework named Text AutoAugment (TAA) to establish a compositional and learnable paradigm for data augmentation. We regard a combination of various operations as an augmentation policy and utilize an efficient Bayesian Optimization algorithm to automatically search for the best policy, which substantially improves the generalization capability of models. Experiments on six benchmark datasets show that TAA boosts classification accuracy in low-resource and class-imbalanced regimes by an average of 8.8% and 9.7%, respectively, outperforming strong baselines.
BibleTTS is a large, high-quality, open speech dataset for ten languages spoken in Sub-Saharan Africa. The corpus contains up to 86 hours of aligned, studio quality 48kHz single speaker recordings per language, enabling the development of high-quality text-to-speech models. The ten languages represented are: Akuapem Twi, Asante Twi, Chichewa, Ewe, Hausa, Kikuyu, Lingala, Luganda, Luo, and Yoruba. This corpus is a derivative work of Bible recordings made and released by the Open.Bible project from Biblica. We have aligned, cleaned, and filtered the original recordings, and additionally hand-checked a subset of the alignments for each language. We present results for text-to-speech models with Coqui TTS. The data is released under a commercial-friendly CC-BY-SA license.
We show that autoregressive language models can learn to infill text after we apply a straightforward transformation to the dataset, which simply moves a span of text from the middle of a document to its end. While this data augmentation has garnered much interest in recent years, we provide extensive evidence that training models with a large fraction of data transformed in this way does not harm the original left-to-right generative capability, as measured by perplexity and sampling evaluations across a wide range of scales. Given the usefulness, simplicity, and efficiency of training models to fill-in-the-middle (FIM), we suggest that future autoregressive language models be trained with FIM by default. To this end, we run a series of ablations on key hyperparameters, such as the data transformation frequency, the structure of the transformation, and the method of selecting the infill span. We use these ablations to prescribe strong default settings and best practices to train FIM models. We have released our best infilling model trained with best practices in our API, and release our infilling benchmarks to aid future research.
We address the problem of cross-speaker style transfer for text-to-speech (TTS) using data augmentation via voice conversion. We assume to have a corpus of neutral non-expressive data from a target speaker and supporting conversational expressive data from different speakers. Our goal is to build a TTS system that is expressive, while retaining the target speaker's identity. The proposed approach relies on voice conversion to first generate high-quality data from the set of supporting expressive speakers. The voice converted data is then pooled with natural data from the target speaker and used to train a single-speaker multi-style TTS system. We provide evidence that this approach is efficient, flexible, and scalable. The method is evaluated using one or more supporting speakers, as well as a variable amount of supporting data. We further provide evidence that this approach allows some controllability of speaking style, when using multiple supporting speakers. We conclude by scaling our proposed technology to a set of 14 speakers across 7 languages. Results indicate that our technology consistently improves synthetic samples in terms of style similarity, while retaining the target speaker's identity.
Real-world recognition system often encounters a plenty of unseen labels in practice. To identify such unseen labels, multi-label zero-shot learning (ML-ZSL) focuses on transferring knowledge by a pre-trained textual label embedding (e.g., GloVe). However, such methods only exploit singlemodal knowledge from a language model, while ignoring the rich semantic information inherent in image-text pairs. Instead, recently developed open-vocabulary (OV) based methods succeed in exploiting such information of image-text pairs in object detection, and achieve impressive performance. Inspired by the success of OV-based methods, we propose a novel open-vocabulary framework, named multimodal knowledge transfer (MKT), for multi-label classification. Specifically, our method exploits multi-modal knowledge of image-text pairs based on a vision and language pretraining (VLP) model. To facilitate transferring the imagetext matching ability of VLP model, knowledge distillation is used to guarantee the consistency of image and label embeddings, along with prompt tuning to further update the label embeddings. To further recognize multiple objects, a simple but effective two-stream module is developed to capture both local and global features. Extensive experimental results show that our method significantly outperforms state-of-theart methods on public benchmark datasets. Code will be available at https://github.com/seanhe97/MKT.
Optical character recognition (OCR) technology has been widely used in various scenes, as shown in Figure 1. Designing a practical OCR system is still a meaningful but challenging task. In previous work, considering the efficiency and accuracy, we proposed a practical ultra lightweight OCR system (PP-OCR), and an optimized version PP-OCRv2. In order to further improve the performance of PP-OCRv2, a more robust OCR system PP-OCRv3 is proposed in this paper. PP-OCRv3 upgrades the text detection model and text recognition model in 9 aspects based on PP-OCRv2. For text detector, we introduce a PAN module with large receptive field named LK-PAN, a FPN module with residual attention mechanism named RSE-FPN, and DML distillation strategy. For text recognizer, the base model is replaced from CRNN to SVTR, and we introduce lightweight text recognition network SVTR LCNet, guided training of CTC by attention, data augmentation strategy TextConAug, better pre-trained model by self-supervised TextRotNet, UDML, and UIM to accelerate the model and improve the effect. Experiments on real data show that the hmean of PP-OCRv3 is 5% higher than PP-OCRv2 under comparable inference speed. All the above mentioned models are open-sourced and the code is available in the GitHub repository PaddleOCR which is powered by PaddlePaddle.
Vocabulary transfer is a transfer learning subtask in which language models fine-tune with the corpus-specific tokenization instead of the default one, which is being used during pretraining. This usually improves the resulting performance of the model, and in the paper, we demonstrate that vocabulary transfer is especially beneficial for medical text processing. Using three different medical natural language processing datasets, we show vocabulary transfer to provide up to ten extra percentage points for the downstream classifier accuracy.
We propose a novel framework for modeling the interaction between graphical structures and the natural language text associated with their nodes and edges. Existing approaches typically fall into two categories. On group ignores the relational structure by converting them into linear sequences and then utilize the highly successful Seq2Seq models. The other side ignores the sequential nature of the text by representing them as fixed-dimensional vectors and apply graph neural networks. Both simplifications lead to information loss. Our proposed method utilizes both the graphical structure as well as the sequential nature of the texts. The input to our model is a set of text segments associated with the nodes and edges of the graph, which are then processed with a transformer encoder-decoder model, equipped with a self-attention mechanism that is aware of the graphical relations between the nodes containing the segments. This also allows us to use BERT-like models that are already trained on large amounts of text. While the proposed model has wide applications, we demonstrate its capabilities on data-to-text generation tasks. Our approach compares favorably against state-of-the-art methods in four tasks without tailoring the model architecture. We also provide an early demonstration in a novel practical application -- generating clinical notes from the medical entities mentioned during clinical visits.