Inspired by the success of visual-language methods (VLMs) in zero-shot classification, recent works attempt to extend this line of work into object detection by leveraging the localization ability of pre-trained VLMs and generating pseudo labels for unseen classes in a self-training manner. However, since the current VLMs are usually pre-trained with aligning sentence embedding with global image embedding, the direct use of them lacks fine-grained alignment for object instances, which is the core of detection. In this paper, we propose a simple but effective Pretrain-adaPt-Pseudo labeling paradigm for Open-Vocabulary Detection (P$^3$OVD) that introduces a fine-grained visual-text prompt adapting stage to enhance the current self-training paradigm with a more powerful fine-grained alignment. During the adapting stage, we enable VLM to obtain fine-grained alignment by using learnable text prompts to resolve an auxiliary dense pixel-wise prediction task. Furthermore, we propose a visual prompt module to provide the prior task information (i.e., the categories need to be predicted) for the vision branch to better adapt the pretrained VLM to the downstream tasks. Experiments show that our method achieves the state-of-the-art performance for open-vocabulary object detection, e.g., 31.5% mAP on unseen classes of COCO.
With the similarity between music and speech synthesis from symbolic input and the rapid development of text-to-speech (TTS) techniques, it is worthwhile to explore ways to improve the MIDI-to-audio performance by borrowing from TTS techniques. In this study, we analyze the shortcomings of a TTS-based MIDI-to-audio system and improve it in terms of feature computation, model selection, and training strategy, aiming to synthesize highly natural-sounding audio. Moreover, we conducted an extensive model evaluation through listening tests, pitch measurement, and spectrogram analysis. This work demonstrates not only synthesis of highly natural music but offers a thorough analytical approach and useful outcomes for the community. Our code and pre-trained models are open sourced at https://github.com/nii-yamagishilab/midi-to-audio.
The lack of data and the difficulty of multimodal fusion have always been challenges for multimodal emotion recognition (MER). In this paper, we propose to use pretrained models as upstream network, wav2vec 2.0 for audio modality and BERT for text modality, and finetune them in downstream task of MER to cope with the lack of data. For the difficulty of multimodal fusion, we use a K-layer multi-head attention mechanism as a downstream fusion module. Starting from the MER task itself, we design two auxiliary tasks to alleviate the insufficient fusion between modalities and guide the network to capture and align emotion-related features. Compared to the previous state-of-the-art models, we achieve a better performance by 78.42% Weighted Accuracy (WA) and 79.71% Unweighted Accuracy (UA) on the IEMOCAP dataset.
Most computer vision and machine learning-based approaches for historical document analysis are tailored to grayscale or RGB images and thus, mostly exploit their spatial information. Multispectral (MS) and hyperspectral (HS) images contain, next to the spatial information, much richer spectral information than RGB images (usually spreading beyond the visible spectral range) that can facilitate more effective feature extraction, more accurate classification and recognition, and thus, improved analysis. Although utilization of rich spectral information can improve historical document analysis tremendously, there are still some potential limitations of HS imagery such as camera-induced noise and blur that require a carefully designed preprocessing step. Here, we propose novel blind HS image deblurring methods tailored to document images. We exploit a low-rank property of HS images (i.e., by projecting an HS image to a lower dimensional subspace) and utilize a text tailor image prior to performing a PSF estimation and deblurring of subspace components. The preliminary results show that the proposed approach gives good results over all spectral bands, removing successfully image artefacts introduced by blur and noise and significantly increasing the number of bands that can be used in further analysis.
We propose a unified framework for low resource automatic speech recognition tasks named meta audio concatenation (MAC). It is easy to implement and can be carried out in extremely low resource environments. Mathematically, we give a clear description of MAC framework from the perspective of bayesian sampling. In this framework, we leverage a novel concatenative synthesis text-to-speech system to boost the low resource ASR task. By the concatenative synthesis text-to-speech system, we can integrate language pronunciation rules and adjust the TTS process. Furthermore, we propose a broad notion of meta audio set to meet the modeling needs of different languages and different scenes when using the system. Extensive experiments have demonstrated the great effectiveness of MAC on low resource ASR tasks. For CTC greedy search, CTC prefix, attention, and attention rescoring decode mode in Cantonese ASR task, Taiwanese ASR task, and Japanese ASR task the MAC method can reduce the CER by more than 15\%. Furthermore, in the ASR task, MAC beats wav2vec2 (with fine-tuning) on common voice datasets of Cantonese and gets really competitive results on common voice datasets of Taiwanese and Japanese. Among them, it is worth mentioning that we achieve a \textbf{10.9\%} character error rate (CER) on the common voice Cantonese ASR task, bringing about \textbf{30\%} relative improvement compared to the wav2vec2 (with fine-tuning).
The 3D-aware image synthesis focuses on conserving spatial consistency besides generating high-resolution images with fine details. Recently, Neural Radiance Field (NeRF) has been introduced for synthesizing novel views with low computational cost and superior performance. While several works investigate a generative NeRF and show remarkable achievement, they cannot handle conditional and continuous feature manipulation in the generation procedure. In this work, we introduce a novel model, called Class-Continuous Conditional Generative NeRF ($\text{C}^{3}$G-NeRF), which can synthesize conditionally manipulated photorealistic 3D-consistent images by projecting conditional features to the generator and the discriminator. The proposed $\text{C}^{3}$G-NeRF is evaluated with three image datasets, AFHQ, CelebA, and Cars. As a result, our model shows strong 3D-consistency with fine details and smooth interpolation in conditional feature manipulation. For instance, $\text{C}^{3}$G-NeRF exhibits a Fr\'echet Inception Distance (FID) of 7.64 in 3D-aware face image synthesis with a $\text{128}^{2}$ resolution. Additionally, we provide FIDs of generated 3D-aware images of each class of the datasets as it is possible to synthesize class-conditional images with $\text{C}^{3}$G-NeRF.
Semantic-driven 3D shape generation aims to generate 3D objects conditioned on text. Previous works face problems with single-category generation, low-frequency 3D details, and requiring a large number of paired datasets for training. To tackle these challenges, we propose a multi-category conditional diffusion model. Specifically, 1) to alleviate the problem of lack of large-scale paired data, we bridge the text, 2D image and 3D shape based on the pre-trained CLIP model, and 2) to obtain the multi-category 3D shape feature, we apply the conditional flow model to generate 3D shape vector conditioned on CLIP embedding. 3) to generate multi-category 3D shape, we employ the hidden-layer diffusion model conditioned on the multi-category shape vector, which greatly reduces the training time and memory consumption.
Multi-label text classification (MLTC) is one of the key tasks in natural language processing. It aims to assign multiple target labels to one document. Due to the uneven popularity of labels, the number of documents per label follows a long-tailed distribution in most cases. It is much more challenging to learn classifiers for data-scarce tail labels than for data-rich head labels. The main reason is that head labels usually have sufficient information, e.g., a large intra-class diversity, while tail labels do not. In response, we propose a Pairwise Instance Relation Augmentation Network (PIRAN) to augment tailed-label documents for balancing tail labels and head labels. PIRAN consists of a relation collector and an instance generator. The former aims to extract the document pairwise relations from head labels. Taking these relations as perturbations, the latter tries to generate new document instances in high-level feature space around the limited given tailed-label instances. Meanwhile, two regularizers (diversity and consistency) are designed to constrain the generation process. The consistency-regularizer encourages the variance of tail labels to be close to head labels and further balances the whole datasets. And diversity-regularizer makes sure the generated instances have diversity and avoids generating redundant instances. Extensive experimental results on three benchmark datasets demonstrate that PIRAN consistently outperforms the SOTA methods, and dramatically improves the performance of tail labels.
Text Generation Models (TGMs) succeed in creating text that matches human language style reasonably well. Detectors that can distinguish between TGM-generated text and human-written ones play an important role in preventing abuse of TGM. In this paper, we describe our pipeline for the two DIALOG-22 RuATD tasks: detecting generated text (binary task) and classification of which model was used to generate text (multiclass task). We achieved 1st place on the binary classification task with an accuracy score of 0.82995 on the private test set and 4th place on the multiclass classification task with an accuracy score of 0.62856 on the private test set. We proposed an ensemble method of different pre-trained models based on the attention mechanism.
Personality is considered one of the most influential research topics in psychology, as it predicts many consequential outcomes such as mental and physical health and explains human behaviour. With the widespread use of social networks as a means of communication, it is becoming increasingly important to develop models that can automatically and accurately read the essence of individuals based solely on their writing. In particular, the convergence of social and computer sciences has led researchers to develop automatic approaches for extracting and studying "hidden" information in textual data on the internet. The nature of this thesis project is highly experimental, and the motivation behind this work is to present detailed analyses on the topic, as currently there are no significant investigations of this kind. The objective is to identify an adequate semantic space that allows for defining the personality of the object to which a certain text refers. The starting point is a dictionary of adjectives that psychological literature defines as markers of the five major personality traits, or Big Five. In this work, we started with the implementation of fully-connected neural networks as a basis for understanding how simple deep learning models can provide information on hidden personality characteristics. Finally, we use a class of distributional algorithms invented in 2013 by Tomas Mikolov, which consists of using a convolutional neural network that learns the contexts of words in an unsupervised way. In this way, we construct an embedding that contains the semantic information on the text, obtaining a kind of "geometry of meaning" in which concepts are translated into linear relationships. With this last experiment, we hypothesize that an individual writing style is largely coupled with their personality traits.