Shi Chen

Chen Shi, Xiangtai Li, Yanran Wu et al.

Representation of semantic context and local details is the essential issue for building modern semantic segmentation models. However, the interrelationship between semantic context and local details is not well explored in previous works. In this paper, we propose a Dynamic Dual Sampling Module (DDSM) to conduct dynamic affinity modeling and propagate semantic context to local details, which yields a more discriminative representation. Specifically, a dynamic sampling strategy is used to sparsely sample representative pixels and channels in the higher layer, forming adaptive compact support for each pixel and channel in the lower layer. The sampled features with high semantics are aggregated according to the affinities and then propagated to detailed lower-layer features, leading to a fine-grained segmentation result with well-preserved boundaries. Experiment results on both Cityscapes and Camvid datasets validate the effectiveness and efficiency of the proposed approach. Code and models will be available at \url{x3https://github.com/Fantasticarl/DDSM}.

Yanran Wu, Xiangtai Li, Chen Shi et al.

In this paper, we propose an effective method for fast and accurate scene parsing called Bidirectional Alignment Network (BiAlignNet). Previously, one representative work BiSeNet~\cite{bisenet} uses two different paths (Context Path and Spatial Path) to achieve balanced learning of semantics and details, respectively. However, the relationship between the two paths is not well explored. We argue that both paths can benefit each other in a complementary way. Motivated by this, we propose a novel network by aligning two-path information into each other through a learned flow field. To avoid the noise and semantic gaps, we introduce a Gated Flow Alignment Module to align both features in a bidirectional way. Moreover, to make the Spatial Path learn more detailed information, we present an edge-guided hard pixel mining loss to supervise the aligned learning process. Our method achieves 80.1\% and 78.5\% mIoU in validation and test set of Cityscapes while running at 30 FPS with full resolution inputs. Code and models will be available at \url{https://github.com/jojacola/BiAlignNet}.

Bo Pang, Kaiwen Zha, Hanwen Cao et al.

Extracting temporal and representation features efficiently plays a pivotal role in understanding visual sequence information. To deal with this, we propose a new recurrent neural framework that can be stacked deep effectively. There are mainly two novel designs in our deep RNN framework: one is a new RNN module called Context Bridge Module (CBM) which splits the information flowing along the sequence (temporal direction) and along depth (spatial representation direction), making it easier to train when building deep by balancing these two directions; the other is the Overlap Coherence Training Scheme that reduces the training complexity for long visual sequential tasks on account of the limitation of computing resources. We provide empirical evidence to show that our deep RNN framework is easy to optimize and can gain accuracy from the increased depth on several visual sequence problems. On these tasks, we evaluate our deep RNN framework with 15 layers, 7* than conventional RNN networks, but it is still easy to train. Our deep framework achieves more than 11% relative improvements over shallow RNN models on Kinetics, UCF-101, and HMDB-51 for video classification. For auxiliary annotation, after replacing the shallow RNN part of Polygon-RNN with our 15-layer deep CBM, the performance improves by 14.7%. For video future prediction, our deep RNN improves the state-of-the-art shallow model's performance by 2.4% on PSNR and SSIM. The code and trained models are published accompanied by this paper: https://github.com/BoPang1996/Deep-RNN-Framework.

Mengyang Feng, Jinlin Liu, Kai Yu et al.

In this paper, we present DreaMoving, a diffusion-based controllable video generation framework to produce high-quality customized human videos. Specifically, given target identity and posture sequences, DreaMoving can generate a video of the target identity moving or dancing anywhere driven by the posture sequences. To this end, we propose a Video ControlNet for motion-controlling and a Content Guider for identity preserving. The proposed model is easy to use and can be adapted to most stylized diffusion models to generate diverse results. The project page is available at https://dreamoving.github.io/dreamoving

Xingyu Song, Zhan Li, Shi Chen et al.

3D human pose estimation is a vital task in computer vision, involving the prediction of human joint positions from images or videos to reconstruct a skeleton of a human in three-dimensional space. This technology is pivotal in various fields, including animation, security, human-computer interaction, and automotive safety, where it promotes both technological progress and enhanced human well-being. The advent of deep learning significantly advances the performance of 3D pose estimation by incorporating temporal information for predicting the spatial positions of human joints. However, traditional methods often fall short as they primarily focus on the spatial coordinates of joints and overlook the orientation and rotation of the connecting bones, which are crucial for a comprehensive understanding of human pose in 3D space. To address these limitations, we introduce Quater-GCN (Q-GCN), a directed graph convolutional network tailored to enhance pose estimation by orientation. Q-GCN excels by not only capturing the spatial dependencies among node joints through their coordinates but also integrating the dynamic context of bone rotations in 2D space. This approach enables a more sophisticated representation of human poses by also regressing the orientation of each bone in 3D space, moving beyond mere coordinate prediction. Furthermore, we complement our model with a semi-supervised training strategy that leverages unlabeled data, addressing the challenge of limited orientation ground truth data. Through comprehensive evaluations, Q-GCN has demonstrated outstanding performance against current state-of-the-art methods.

Ziang Chen, Zhengjiang Lin, Shi Chen et al.

Graph neural networks (GNNs) have achieved remarkable empirical success in processing and representing graph-structured data across various domains. However, a significant challenge known as "oversmoothing" persists, where vertex features become nearly indistinguishable in deep GNNs, severely restricting their expressive power and practical utility. In this work, we analyze the asymptotic oversmoothing rates of deep GNNs with and without residual connections by deriving explicit convergence rates for a normalized vertex similarity measure. Our analytical framework is grounded in the multiplicative ergodic theorem. Furthermore, we demonstrate that adding residual connections effectively mitigates or prevents oversmoothing across several broad families of parameter distributions. The theoretical findings are strongly supported by numerical experiments.

Xingyu Song, Zhan Li, Shi Chen et al.

Action recognition, an essential component of computer vision, plays a pivotal role in multiple applications. Despite significant improvements brought by Convolutional Neural Networks (CNNs), these models suffer performance declines when trained with discontinuous video frames, which is a frequent scenario in real-world settings. This decline primarily results from the loss of temporal continuity, which is crucial for understanding the semantics of human actions. To overcome this issue, we introduce the 4A (Action Animation-based Augmentation Approach) pipeline, which employs a series of sophisticated techniques: starting with 2D human pose estimation from RGB videos, followed by Quaternion-based Graph Convolution Network for joint orientation and trajectory prediction, and Dynamic Skeletal Interpolation for creating smoother, diversified actions using game engine technology. This innovative approach generates realistic animations in varied game environments, viewed from multiple viewpoints. In this way, our method effectively bridges the domain gap between virtual and real-world data. In experimental evaluations, the 4A pipeline achieves comparable or even superior performance to traditional training approaches using real-world data, while requiring only 10% of the original data volume. Additionally, our approach demonstrates enhanced performance on In-the-wild videos, marking a significant advancement in the field of action recognition.

Xingyu Song, Zhan Li, Shi Chen et al.

Current datasets for action recognition tasks face limitations stemming from traditional collection and generation methods, including the constrained range of action classes, absence of multi-viewpoint recordings, limited diversity, poor video quality, and labor-intensive manually collection. To address these challenges, we introduce GTAutoAct, a innovative dataset generation framework leveraging game engine technology to facilitate advancements in action recognition. GTAutoAct excels in automatically creating large-scale, well-annotated datasets with extensive action classes and superior video quality. Our framework's distinctive contributions encompass: (1) it innovatively transforms readily available coordinate-based 3D human motion into rotation-orientated representation with enhanced suitability in multiple viewpoints; (2) it employs dynamic segmentation and interpolation of rotation sequences to create smooth and realistic animations of action; (3) it offers extensively customizable animation scenes; (4) it implements an autonomous video capture and processing pipeline, featuring a randomly navigating camera, with auto-trimming and labeling functionalities. Experimental results underscore the framework's robustness and highlights its potential to significantly improve action recognition model training.

Zihan Liu, Feijun Jiang, Yuxiang Hu et al.

Named entity recognition (NER) models generally perform poorly when large training datasets are unavailable for low-resource domains. Recently, pre-training a large-scale language model has become a promising direction for coping with the data scarcity issue. However, the underlying discrepancies between the language modeling and NER task could limit the models' performance, and pre-training for the NER task has rarely been studied since the collected NER datasets are generally small or large but with low quality. In this paper, we construct a massive NER corpus with a relatively high quality, and we pre-train a NER-BERT model based on the created dataset. Experimental results show that our pre-trained model can significantly outperform BERT as well as other strong baselines in low-resource scenarios across nine diverse domains. Moreover, a visualization of entity representations further indicates the effectiveness of NER-BERT for categorizing a variety of entities.

Akihiro Nakano, Shi Chen, Kazuyuki Demachi

Multi-task learning (MTL) is an active field in deep learning in which we train a model to jointly learn multiple tasks by exploiting relationships between the tasks. It has been shown that MTL helps the model share the learned features between tasks and enhance predictions compared to when learning each task independently. We propose a new learning framework for 2-task MTL problem that uses the predictions of one task as inputs to another network to predict the other task. We define two new loss terms inspired by cycle-consistency loss and contrastive learning, alignment loss and cross-task consistency loss. Both losses are designed to enforce the model to align the predictions of multiple tasks so that the model predicts consistently. We theoretically prove that both losses help the model learn more efficiently and that cross-task consistency loss is better in terms of alignment with the straight-forward predictions. Experimental results also show that our proposed model achieves significant performance on the benchmark Cityscapes and NYU dataset.

Zhaojiang Lin, Andrea Madotto, Genta Indra Winata et al.

Task-oriented dialogue (ToD) benchmarks provide an important avenue to measure progress and develop better conversational agents. However, existing datasets for end-to-end ToD modeling are limited to a single language, hindering the development of robust end-to-end ToD systems for multilingual countries and regions. Here we introduce BiToD, the first bilingual multi-domain dataset for end-to-end task-oriented dialogue modeling. BiToD contains over 7k multi-domain dialogues (144k utterances) with a large and realistic bilingual knowledge base. It serves as an effective benchmark for evaluating bilingual ToD systems and cross-lingual transfer learning approaches. We provide state-of-the-art baselines under three evaluation settings (monolingual, bilingual, and cross-lingual). The analysis of our baselines in different settings highlights 1) the effectiveness of training a bilingual ToD system compared to two independent monolingual ToD systems, and 2) the potential of leveraging a bilingual knowledge base and cross-lingual transfer learning to improve the system performance under low resource condition.

Zhiyan Ding, Shi Chen, Qin Li et al.

Finding parameters in a deep neural network (NN) that fit training data is a nonconvex optimization problem, but a basic first-order optimization method (gradient descent) finds a global optimizer with perfect fit (zero-loss) in many practical situations. We examine this phenomenon for the case of Residual Neural Networks (ResNet) with smooth activation functions in a limiting regime in which both the number of layers (depth) and the number of weights in each layer (width) go to infinity. First, we use a mean-field-limit argument to prove that the gradient descent for parameter training becomes a gradient flow for a probability distribution that is characterized by a partial differential equation (PDE) in the large-NN limit. Next, we show that under certain assumptions, the solution to the PDE converges in the training time to a zero-loss solution. Together, these results suggest that the training of the ResNet gives a near-zero loss if the ResNet is large enough. We give estimates of the depth and width needed to reduce the loss below a given threshold, with high probability.