Jinsong Zhang

Bobo Li, Hao Fei, Fei Li et al.

The rapid development of aspect-based sentiment analysis (ABSA) within recent decades shows great potential for real-world society. The current ABSA works, however, are mostly limited to the scenario of a single text piece, leaving the study in dialogue contexts unexplored. To bridge the gap between fine-grained sentiment analysis and conversational opinion mining, in this work, we introduce a novel task of conversational aspect-based sentiment quadruple analysis, namely DiaASQ, aiming to detect the quadruple of target-aspect-opinion-sentiment in a dialogue. We manually construct a large-scale high-quality DiaASQ dataset in both Chinese and English languages. We deliberately develop a neural model to benchmark the task, which advances in effectively performing end-to-end quadruple prediction, and manages to incorporate rich dialogue-specific and discourse feature representations for better cross-utterance quadruple extraction. We hope the new benchmark will spur more advancements in the sentiment analysis community.

Haibiao Xuan, Xiongzheng Li, Jinsong Zhang et al.

Naturally controllable human-scene interaction (HSI) generation has an important role in various fields, such as VR/AR content creation and human-centered AI. However, existing methods are unnatural and unintuitive in their controllability, which heavily limits their application in practice. Therefore, we focus on a challenging task of naturally and controllably generating realistic and diverse HSIs from textual descriptions. From human cognition, the ideal generative model should correctly reason about spatial relationships and interactive actions. To that end, we propose Narrator, a novel relationship reasoning-based generative approach using a conditional variation autoencoder for naturally controllable generation given a 3D scene and a textual description. Also, we model global and local spatial relationships in a 3D scene and a textual description respectively based on the scene graph, and introduce a partlevel action mechanism to represent interactions as atomic body part states. In particular, benefiting from our relationship reasoning, we further propose a simple yet effective multi-human generation strategy, which is the first exploration for controllable multi-human scene interaction generation. Our extensive experiments and perceptual studies show that Narrator can controllably generate diverse interactions and significantly outperform existing works. The code and dataset will be available for research purposes.

Jinsong Zhang, Minjie Zhu, Yuxiang Zhang et al.

This paper addresses the problem of generating whole-body motion from speech. Despite great successes, prior methods still struggle to produce reasonable and diverse whole-body motions from speech. This is due to their reliance on suboptimal representations and a lack of strategies for generating diverse results. To address these challenges, we present a novel hybrid point representation to achieve accurate and continuous motion generation, e.g., avoiding foot skating, and this representation can be transformed into an easy-to-use representation, i.e., SMPL-X body mesh, for many applications. To generate whole-body motion from speech, for facial motion, closely tied to the audio signal, we introduce an encoder-decoder architecture to achieve deterministic outcomes. However, for the body and hands, which have weaker connections to the audio signal, we aim to generate diverse yet reasonable motions. To boost diversity in motion generation, we propose a contrastive motion learning method to encourage the model to produce more distinctive representations. Specifically, we design a robust VQ-VAE to learn a quantized motion codebook using our hybrid representation. Then, we regress the motion representation from the audio signal by a translation model employing our contrastive motion learning method. Experimental results validate the superior performance and the correctness of our model. The project page is available for research purposes at http://cic.tju.edu.cn/faculty/likun/projects/SpeechAct.

Xiongzheng Li, Jinsong Zhang, Yu-Kun Lai et al.

Much progress has been made in reconstructing garments from an image or a video. However, none of existing works meet the expectations of digitizing high-quality animatable dynamic garments that can be adjusted to various unseen poses. In this paper, we propose the first method to recover high-quality animatable dynamic garments from monocular videos without depending on scanned data. To generate reasonable deformations for various unseen poses, we propose a learnable garment deformation network that formulates the garment reconstruction task as a pose-driven deformation problem. To alleviate the ambiguity estimating 3D garments from monocular videos, we design a multi-hypothesis deformation module that learns spatial representations of multiple plausible deformations. Experimental results on several public datasets demonstrate that our method can reconstruct high-quality dynamic garments with coherent surface details, which can be easily animated under unseen poses. The code will be provided for research purposes.

Linkai Peng, Yingming Gao, Binghuai Lin et al.

Mispronunciation detection and diagnosis (MDD) technology is a key component of computer-assisted pronunciation training system (CAPT). In the field of assessing the pronunciation quality of constrained speech, the given transcriptions can play the role of a teacher. Conventional methods have fully utilized the prior texts for the model construction or improving the system performance, e.g. forced-alignment and extended recognition networks. Recently, some end-to-end based methods attempt to incorporate the prior texts into model training and preliminarily show the effectiveness. However, previous studies mostly consider applying raw attention mechanism to fuse audio representations with text representations, without taking possible text-pronunciation mismatch into account. In this paper, we present a gating strategy that assigns more importance to the relevant audio features while suppressing irrelevant text information. Moreover, given the transcriptions, we design an extra contrastive loss to reduce the gap between the learning objective of phoneme recognition and MDD. We conducted experiments using two publicly available datasets (TIMIT and L2-Arctic) and our best model improved the F1 score from $57.51\%$ to $61.75\%$ compared to the baselines. Besides, we provide a detailed analysis to shed light on the effectiveness of gating mechanism and contrastive learning on MDD.

Yuqi Shi, Hao Yang, Xiyao Lu et al.

While second language (L2) learners may acquire target syntactic word order, mapping this syntax onto appropriate prosodic structures remains a persistent challenge. This study investigates the fossilization and stability of the L2 syntax-prosody interface by comparing 67 native Mandarin speakers with 67 Vietnamese learners using the BLCU-SAIT corpus. By integrating C-ToBI boundary annotation with Dependency Grammar analysis, we examined both the quantity of prosodic boundaries and their mapping to syntactic relations. Results reveal a non-linear acquisition: although high-proficiency learners (VNH) converge to the native baseline in boundary quantity at the Major Phrase level (B3), their structural mapping significantly diverges. Specifically, VNH demote the prosodic boundary at the Subject-Verb (SBV) interface (Major Phrase B3 -> Prosodic Word B1), while erroneously promoting the boundary at the Verb-Object (VOB) interface (Prosodic Word B1 -> Major Phrase B3). This strategy allows learners to maintain high long phrasal output at the expense of structural accuracy. This results in a distorted prosodic hierarchy where the native pattern is inverted.

Hailong Chu, Hongbing Li, Yunlong Chu et al.

Multimedia event extraction (M2E2) aims to predict triggers, ground arguments across text and images, and then assemble them into schema-consistent event records. Recent LLM-based approaches have shown strong potential for M2E2, but their intermediate event hypotheses often remain implicit, and event-argument linking is still tightly coupled with role binding. This leaves little opportunity to inspect or revise intermediate event hypotheses and makes predictions brittle to early errors. To bridge this gap, we present ECHO, a multi-agent framework that reframes M2E2 as iterative refinement over an explicit Multimedia Event Hypergraph (MEHG). Instead of relying on implicit linear generation, ECHO performs auditable atomic updates over a shared hypergraph, making intermediate event structures explicit and revisable. Furthermore, we introduce a Link-then-Bind strategy that decouples event-argument linking from role binding, reducing premature semantic commitment during structured prediction. Extensive experiments on the M2E2 benchmark show that ECHO consistently outperforms prior state-of-the-art approaches, achieving gains of 7.3 and 15.5 F1 points on event mention and argument role, respectively.

Jinsong Zhang, Minghe Li, Jiayi Tian et al.

High-order tensor decomposition has been widely adopted to obtain compact deep neural networks for edge deployment. However, existing studies focus primarily on its algorithmic advantages such as accuracy and compression ratio-while overlooking the hardware deployment efficiency. Such hardware-unaware designs often obscure the potential latency and energy benefits of tensorized models. Although several works attempt to reduce computational cost by optimizing the contraction sequence based on the number of multiply-accumulate operations, they typically neglect the underlying hardware characteristics, resulting in suboptimal real-world performance. We observe that the contraction path, hardware architecture, and dataflow mapping are tightly coupled and must be optimized jointly within a unified design space to maximize deployment efficiency on real devices. To this end, we propose a co-exploration framework that unifies these dimensions within a unified design space for efficient training and inference of tensorized neural networks on edge platforms. The framework formulates a latency oriented search objective and solves it via a global latency-driven exploration across the unified design space to achieve end-to-end model efficiency. The optimized configurations are implemented on a configurable FPGA kernel, achieving up to 4x and 3.85x lower inference and training latency compared with the dense baseline.

Shuo Zhang, Jinsong Zhang, Zhejun Zhang et al.

Multi-task learning (MTL) enables the efficient transfer of extra knowledge acquired from other tasks. The high correlation between multimodal sentiment analysis (MSA) and multimodal emotion recognition (MER) supports their joint training. However, existing methods primarily employ hard parameter sharing, ignoring parameter conflicts caused by complex task correlations. In this paper, we present a novel MTL method for MSA and MER, termed Multimodal Mixture of Low-Rank Experts (MMoLRE). MMoLRE utilizes shared and task-specific experts to distinctly model common and unique task characteristics, thereby avoiding parameter conflicts. Additionally, inspired by low-rank structures in the Mixture of Experts (MoE) framework, we design low-rank expert networks to reduce parameter and computational overhead as the number of experts increases. Extensive experiments on the CMU-MOSI and CMU-MOSEI benchmarks demonstrate that MMoLRE achieves state-of-the-art performance on the MSA task and competitive results on the MER task.

Jinsong Zhang, Lingfeng Gu, Yu-Kun Lai et al.

Group detection, especially for large-scale scenes, has many potential applications for public safety and smart cities. Existing methods fail to cope with frequent occlusions in large-scale scenes with multiple people, and are difficult to effectively utilize spatio-temporal information. In this paper, we propose an end-to-end framework,GroupTransformer, for group detection in large-scale scenes. To deal with the frequent occlusions caused by multiple people, we design an occlusion encoder to detect and suppress severely occluded person crops. To explore the potential spatio-temporal relationship, we propose spatio-temporal transformers to simultaneously extract trajectory information and fuse inter-person features in a hierarchical manner. Experimental results on both large-scale and small-scale scenes demonstrate that our method achieves better performance compared with state-of-the-art methods. On large-scale scenes, our method significantly boosts the performance in terms of precision and F1 score by more than 10%. On small-scale scenes, our method still improves the performance of F1 score by more than 5%. The project page with code can be found at http://cic.tju.edu.cn/faculty/likun/projects/GroupTrans.

Huahuan Zheng, Wenjie Peng, Zhijian Ou et al.

Automatic speech recognition systems have been largely improved in the past few decades and current systems are mainly hybrid-based and end-to-end-based. The recently proposed CTC-CRF framework inherits the data-efficiency of the hybrid approach and the simplicity of the end-to-end approach. In this paper, we further advance CTC-CRF based ASR technique with explorations on modeling units and neural architectures. Specifically, we investigate techniques to enable the recently developed wordpiece modeling units and Conformer neural networks to be succesfully applied in CTC-CRFs. Experiments are conducted on two English datasets (Switchboard, Librispeech) and a German dataset from CommonVoice. Experimental results suggest that (i) Conformer can improve the recognition performance significantly; (ii) Wordpiece-based systems perform slightly worse compared with phone-based systems for the target language with a low degree of grapheme-phoneme correspondence (e.g. English), while the two systems can perform equally strong when such degree of correspondence is high for the target language (e.g. German).

Dengfeng Ke, Jinsong Zhang, Yanlu Xie et al.

Single channel speech enhancement is a challenging task in speech community. Recently, various neural networks based methods have been applied to speech enhancement. Among these models, PHASEN and T-GSA achieve state-of-the-art performances on the publicly opened VoiceBank+DEMAND corpus. Both of the models reach the COVL score of 3.62. PHASEN achieves the highest CSIG score of 4.21 while T-GSA gets the highest PESQ score of 3.06. However, both of these two models are very large. The contradiction between the model performance and the model size is hard to reconcile. In this paper, we introduce three kinds of techniques to shrink the PHASEN model and improve the performance. Firstly, seperable polling attention is proposed to replace the frequency transformation blocks in PHASEN. Secondly, global layer normalization followed with PReLU is used to replace batch normalization followed with ReLU. Finally, BLSTM in PHASEN is replaced with Conv2d operation and the phase stream is simplified. With all these modifications, the size of the PHASEN model is shrunk from 33M parameters to 5M parameters, while the performance on VoiceBank+DEMAND is improved to the CSIG score of 4.30, the PESQ score of 3.07 and the COVL score of 3.73.

Kaiqi Fu, Jones Lin, Dengfeng Ke et al.

Recently, end-to-end mispronunciation detection and diagnosis (MD&D) systems has become a popular alternative to greatly simplify the model-building process of conventional hybrid DNN-HMM systems by representing complicated modules with a single deep network architecture. In this paper, in order to utilize the prior text in the end-to-end structure, we present a novel text-dependent model which is difference with sed-mdd, the model achieves a fully end-to-end system by aligning the audio with the phoneme sequences of the prior text inside the model through the attention mechanism. Moreover, the prior text as input will be a problem of imbalance between positive and negative samples in the phoneme sequence. To alleviate this problem, we propose three simple data augmentation methods, which effectively improve the ability of model to capture mispronounced phonemes. We conduct experiments on L2-ARCTIC, and our best performance improved from 49.29% to 56.08% in F-measure metric compared to the CNN-RNN-CTC model.

Jinsong Zhang, Kun Li, Yu-Kun Lai et al.

Person image synthesis, e.g., pose transfer, is a challenging problem due to large variation and occlusion. Existing methods have difficulties predicting reasonable invisible regions and fail to decouple the shape and style of clothing, which limits their applications on person image editing. In this paper, we propose PISE, a novel two-stage generative model for Person Image Synthesis and Editing, which is able to generate realistic person images with desired poses, textures, or semantic layouts. For human pose transfer, we first synthesize a human parsing map aligned with the target pose to represent the shape of clothing by a parsing generator, and then generate the final image by an image generator. To decouple the shape and style of clothing, we propose joint global and local per-region encoding and normalization to predict the reasonable style of clothing for invisible regions. We also propose spatial-aware normalization to retain the spatial context relationship in the source image. The results of qualitative and quantitative experiments demonstrate the superiority of our model on human pose transfer. Besides, the results of texture transfer and region editing show that our model can be applied to person image editing.

Kun Li, Jinsong Zhang, Yebin Liu et al.

Human pose transfer, which aims at transferring the appearance of a given person to a target pose, is very challenging and important in many applications. Previous work ignores the guidance of pose features or only uses local attention mechanism, leading to implausible and blurry results. We propose a new human pose transfer method using a generative adversarial network (GAN) with simplified cascaded blocks. In each block, we propose a pose-guided non-local attention (PoNA) mechanism with a long-range dependency scheme to select more important regions of image features to transfer. We also design pre-posed image-guided pose feature update and post-posed pose-guided image feature update to better utilize the pose and image features. Our network is simple, stable, and easy to train. Quantitative and qualitative results on Market-1501 and DeepFashion datasets show the efficacy and efficiency of our model. Compared with state-of-the-art methods, our model generates sharper and more realistic images with rich details, while having fewer parameters and faster speed. Furthermore, our generated images can help to alleviate data insufficiency for person re-identification.

Jinsong Zhang, Xingzi Liu, Kun Li

Human pose transfer, as a misaligned image generation task, is very challenging. Existing methods cannot effectively utilize the input information, which often fail to preserve the style and shape of hair and clothes. In this paper, we propose an adaptive human pose transfer network with two hierarchical deformation levels. The first level generates human semantic parsing aligned with the target pose, and the second level generates the final textured person image in the target pose with the semantic guidance. To avoid the drawback of vanilla convolution that treats all the pixels as valid information, we use gated convolution in both two levels to dynamically select the important features and adaptively deform the image layer by layer. Our model has very few parameters and is fast to converge. Experimental results demonstrate that our model achieves better performance with more consistent hair, face and clothes with fewer parameters than state-of-the-art methods. Furthermore, our method can be applied to clothing texture transfer.

Kun Li, Jing Yang, Nianhong Jiao et al.

3D face reconstruction from a single image is a challenging problem, especially under partial occlusions and extreme poses. This is because the uncertainty of the estimated 2D landmarks will affect the quality of face reconstruction. In this paper, we propose a novel joint 2D and 3D optimization method to adaptively reconstruct 3D face shapes from a single image, which combines the depths of 3D landmarks to solve the uncertain detections of invisible landmarks. The strategy of our method involves two aspects: a coarse-to-fine pose estimation using both 2D and 3D landmarks, and an adaptive 2D and 3D re-weighting based on the refined pose parameter to recover accurate 3D faces. Experimental results on multiple datasets demonstrate that our method can generate high-quality reconstruction from a single color image and is robust for self-occlusion and large poses.

Wang Dai, Jinsong Zhang, Yingming Gao et al.

Formant tracking is one of the most fundamental problems in speech processing. Traditionally, formants are estimated using signal processing methods. Recent studies showed that generic convolutional architectures can outperform recurrent networks on temporal tasks such as speech synthesis and machine translation. In this paper, we explored the use of Temporal Convolutional Network (TCN) for formant tracking. In addition to the conventional implementation, we modified the architecture from three aspects. First, we turned off the "causal" mode of dilated convolution, making the dilated convolution see the future speech frames. Second, each hidden layer reused the output information from all the previous layers through dense connection. Third, we also adopted a gating mechanism to alleviate the problem of gradient disappearance by selectively forgetting unimportant information. The model was validated on the open access formant database VTR. The experiment showed that our proposed model was easy to converge and achieved an overall mean absolute percent error (MAPE) of 8.2% on speech-labeled frames, compared to three competitive baselines of 9.4% (LSTM), 9.1% (Bi-LSTM) and 8.9% (TCN).

Jinsong Zhang, Kalyan Sunkavalli, Yannick Hold-Geoffroy et al.

We present a neural network that predicts HDR outdoor illumination from a single LDR image. At the heart of our work is a method to accurately learn HDR lighting from LDR panoramas under any weather condition. We achieve this by training another CNN (on a combination of synthetic and real images) to take as input an LDR panorama, and regress the parameters of the Lalonde-Matthews outdoor illumination model. This model is trained such that it a) reconstructs the appearance of the sky, and b) renders the appearance of objects lit by this illumination. We use this network to label a large-scale dataset of LDR panoramas with lighting parameters and use them to train our single image outdoor lighting estimation network. We demonstrate, via extensive experiments, that both our panorama and single image networks outperform the state of the art, and unlike prior work, are able to handle weather conditions ranging from fully sunny to overcast skies.

Jinsong Zhang, Rodrigo Verschae, Shohei Nobuhara et al.

Predicting the short-term power output of a photovoltaic panel is an important task for the efficient management of smart grids. Short-term forecasting at the minute scale, also known as nowcasting, can benefit from sky images captured by regular cameras and installed close to the solar panel. However, estimating the weather conditions from these images---sun intensity, cloud appearance and movement, etc.---is a very challenging task that the community has yet to solve with traditional computer vision techniques. In this work, we propose to learn the relationship between sky appearance and the future photovoltaic power output using deep learning. We train several variants of convolutional neural networks which take historical photovoltaic power values and sky images as input and estimate photovoltaic power in a very short term future. In particular, we compare three different architectures based on: a multi-layer perceptron (MLP), a convolutional neural network (CNN), and a long short term memory (LSTM) module. We evaluate our approach quantitatively on a dataset of photovoltaic power values and corresponding images gathered in Kyoto, Japan. Our experiments reveal that the MLP network, already used similarly in previous work, achieves an RMSE skill score of 7% over the commonly-used persistence baseline on the 1-minute future photovoltaic power prediction task. Our CNN-based network improves upon this with a 12% skill score. In contrast, our LSTM-based model, which can learn the temporal dependencies in the data, achieves a 21% RMSE skill score, thus outperforming all other approaches.

Jinsong Zhang, Jean-François Lalonde

Outdoor lighting has extremely high dynamic range. This makes the process of capturing outdoor environment maps notoriously challenging since special equipment must be used. In this work, we propose an alternative approach. We first capture lighting with a regular, LDR omnidirectional camera, and aim to recover the HDR after the fact via a novel, learning-based inverse tonemapping method. We propose a deep autoencoder framework which regresses linear, high dynamic range data from non-linear, saturated, low dynamic range panoramas. We validate our method through a wide set of experiments on synthetic data, as well as on a novel dataset of real photographs with ground truth. Our approach finds applications in a variety of settings, ranging from outdoor light capture to image matching.