Zhenyu Wu

Zhenyu Wu, Xiuwei Xu, Ziwei Wang et al.

In this paper, we propose a novel network framework for indoor 3D object detection to handle variable input frame numbers in practical scenarios. Existing methods only consider fixed frames of input data for a single detector, such as monocular RGB-D images or point clouds reconstructed from dense multi-view RGB-D images. While in practical application scenes such as robot navigation and manipulation, the raw input to the 3D detectors is the RGB-D images with variable frame numbers instead of the reconstructed scene point cloud. However, the previous approaches can only handle fixed frame input data and have poor performance with variable frame input. In order to facilitate 3D object detection methods suitable for practical tasks, we present a novel 3D detection framework named AnyView for our practical applications, which generalizes well across different numbers of input frames with a single model. To be specific, we propose a geometric learner to mine the local geometric features of each input RGB-D image frame and implement local-global feature interaction through a designed spatial mixture module. Meanwhile, we further utilize a dynamic token strategy to adaptively adjust the number of extracted features for each frame, which ensures consistent global feature density and further enhances the generalization after fusion. Extensive experiments on the ScanNet dataset show our method achieves both great generalizability and high detection accuracy with a simple and clean architecture containing a similar amount of parameters with the baselines.

Jialin Guo, Zhenyu Wu, Zhiqiang Zhan et al.

The calibration for deep neural networks is currently receiving widespread attention and research. Miscalibration usually leads to overconfidence of the model. While, under the condition of long-tailed distribution of data, the problem of miscalibration is more prominent due to the different confidence levels of samples in minority and majority categories, and it will result in more serious overconfidence. To address this problem, some current research have designed diverse temperature coefficients for different categories based on temperature scaling (TS) method. However, in the case of rare samples in minority classes, the temperature coefficient is not generalizable, and there is a large difference between the temperature coefficients of the training set and the validation set. To solve this challenge, this paper proposes a dual-branch temperature scaling calibration model (Dual-TS), which considers the diversities in temperature parameters of different categories and the non-generalizability of temperature parameters for rare samples in minority classes simultaneously. Moreover, we noticed that the traditional calibration evaluation metric, Excepted Calibration Error (ECE), gives a higher weight to low-confidence samples in the minority classes, which leads to inaccurate evaluation of model calibration. Therefore, we also propose Equal Sample Bin Excepted Calibration Error (Esbin-ECE) as a new calibration evaluation metric. Through experiments, we demonstrate that our model yields state-of-the-art in both traditional ECE and Esbin-ECE metrics.

Zhenyu Wu, Duc Hoang, Shih-Yao Lin et al.

Estimating the 3D hand pose from a monocular RGB image is important but challenging. A solution is training on large-scale RGB hand images with accurate 3D hand keypoint annotations. However, it is too expensive in practice. Instead, we have developed a learning-based approach to synthesize realistic, diverse, and 3D pose-preserving hand images under the guidance of 3D pose information. We propose a 3D-aware multi-modal guided hand generative network (MM-Hand), together with a novel geometry-based curriculum learning strategy. Our extensive experimental results demonstrate that the 3D-annotated images generated by MM-Hand qualitatively and quantitatively outperform existing options. Moreover, the augmented data can consistently improve the quantitative performance of the state-of-the-art 3D hand pose estimators on two benchmark datasets. The code will be available at https://github.com/ScottHoang/mm-hand.

Zhenyu Wu, Karthik Suresh, Priya Narayanan et al.

Object detection from images captured by Unmanned Aerial Vehicles (UAVs) is becoming increasingly useful. Despite the great success of the generic object detection methods trained on ground-to-ground images, a huge performance drop is observed when they are directly applied to images captured by UAVs. The unsatisfactory performance is owing to many UAV-specific nuisances, such as varying flying altitudes, adverse weather conditions, dynamically changing viewing angles, etc. Those nuisances constitute a large number of fine-grained domains, across which the detection model has to stay robust. Fortunately, UAVs will record meta-data that depict those varying attributes, which are either freely available along with the UAV images, or can be easily obtained. We propose to utilize those free meta-data in conjunction with associated UAV images to learn domain-robust features via an adversarial training framework dubbed Nuisance Disentangled Feature Transform (NDFT), for the specific challenging problem of object detection in UAV images, achieving a substantial gain in robustness to those nuisances. We demonstrate the effectiveness of our proposed algorithm, by showing state-of-the-art performance (single model) on two existing UAV-based object detection benchmarks. The code is available at https://github.com/TAMU-VITA/UAV-NDFT.

Junru Wu, Yue Wang, Zhenyu Wu et al.

The current trend of pushing CNNs deeper with convolutions has created a pressing demand to achieve higher compression gains on CNNs where convolutions dominate the computation and parameter amount (e.g., GoogLeNet, ResNet and Wide ResNet). Further, the high energy consumption of convolutions limits its deployment on mobile devices. To this end, we proposed a simple yet effective scheme for compressing convolutions though applying k-means clustering on the weights, compression is achieved through weight-sharing, by only recording $K$ cluster centers and weight assignment indexes. We then introduced a novel spectrally relaxed $k$-means regularization, which tends to make hard assignments of convolutional layer weights to $K$ learned cluster centers during re-training. We additionally propose an improved set of metrics to estimate energy consumption of CNN hardware implementations, whose estimation results are verified to be consistent with previously proposed energy estimation tool extrapolated from actual hardware measurements. We finally evaluated Deep $k$-Means across several CNN models in terms of both compression ratio and energy consumption reduction, observing promising results without incurring accuracy loss. The code is available at https://github.com/Sandbox3aster/Deep-K-Means

Zhenyu Wu, Fengmao Lv, Chenglizhao Chen et al.

Colorectal polyp segmentation (CPS), an essential problem in medical image analysis, has garnered growing research attention. Recently, the deep learning-based model completely overwhelmed traditional methods in the field of CPS, and more and more deep CPS methods have emerged, bringing the CPS into the deep learning era. To help the researchers quickly grasp the main techniques, datasets, evaluation metrics, challenges, and trending of deep CPS, this paper presents a systematic and comprehensive review of deep-learning-based CPS methods from 2014 to 2023, a total of 115 technical papers. In particular, we first provide a comprehensive review of the current deep CPS with a novel taxonomy, including network architectures, level of supervision, and learning paradigm. More specifically, network architectures include eight subcategories, the level of supervision comprises six subcategories, and the learning paradigm encompasses 12 subcategories, totaling 26 subcategories. Then, we provided a comprehensive analysis the characteristics of each dataset, including the number of datasets, annotation types, image resolution, polyp size, contrast values, and polyp location. Following that, we summarized CPS's commonly used evaluation metrics and conducted a detailed analysis of 40 deep SOTA models, including out-of-distribution generalization and attribute-based performance analysis. Finally, we discussed deep learning-based CPS methods' main challenges and opportunities.

Zhizhou Wang, Jianli Wang, Ruijing Zheng et al.

Low-rank tensor representation (LRTR) has emerged as a powerful tool for multi-dimensional data processing. However, classical LRTR-based methods face two critical limitations: (1) they typically assume that the holistic data is low-rank, this assumption is often violated in real-world scenarios with significant spatial variations; and (2) they are constrained to discrete meshgrid data, limiting their flexibility and applicability. To overcome these limitations, we propose a Superpixel-informed Continuous low-rank Tensor Representation (SCTR) framework, which enables continuous and flexible modeling of multi-dimensional data beyond traditional grid-based constraints. Our approach introduces two main innovations: First, motivated by the observation that semantically coherent regions exhibit stronger low-rank characteristics than holistic data, we employ superpixels as the basic modeling units. This design not only encodes rich semantic information, but also enhances adaptability to diverse forms of data streams. Second, we propose a novel asymmetric low-rank tensor factorization (ALTF) where superpixel-specific factor matrices are parameterized by a shared neural network with specialized heads. By strategically separating global pattern learning from local adaptation, this framework efficiently captures both cross-superpixel commonalities and within-superpixel variations. This yields a representation that is both highly expressive and compact, balancing model efficiency with adaptability. Extensive experiments on several benchmark datasets demonstrate that SCTR achieves 3-5 dB PSNR improvements over existing LRTR-based methods across multispectral images, videos, and color images.

Linjie Li, Zhenyu Wu, Jiaming Liu et al.

Class-incremental learning is dedicated to the development of deep learning models that are capable of acquiring new knowledge while retaining previously learned information. Most methods focus on balanced data distribution for each task, overlooking real-world long-tailed distributions. Therefore, Long-Tailed Class-Incremental Learning has been introduced, which trains on data where head classes have more samples than tail classes. Existing methods mainly focus on preserving representative samples from previous classes to combat catastrophic forgetting. Recently, dynamic network algorithms freeze old network structures and expand new ones, achieving significant performance. However, with the introduction of the long-tail problem, merely extending Determined blocks can lead to miscalibrated predictions, while expanding the entire backbone results in an explosion of memory size. To address these issues, we introduce a novel Task-aware Expandable (TaE) framework, dynamically allocating and updating task-specific trainable parameters to learn diverse representations from each incremental task while resisting forgetting through the majority of frozen model parameters. To further encourage the class-specific feature representation, we develop a Centroid-Enhanced (CEd) method to guide the update of these task-aware parameters. This approach is designed to adaptively allocate feature space for every class by adjusting the distance between intra- and inter-class features, which can extend to all "training from sketch" algorithms. Extensive experiments demonstrate that TaE achieves state-of-the-art performance.

Yuanhao Zhai, Ziyi Liu, Zhenyu Wu et al.

Deep learning models have a risk of utilizing spurious clues to make predictions, such as recognizing actions based on the background scene. This issue can severely degrade the open-set action recognition performance when the testing samples have different scene distributions from the training samples. To mitigate this problem, we propose a novel method, called Scene-debiasing Open-set Action Recognition (SOAR), which features an adversarial scene reconstruction module and an adaptive adversarial scene classification module. The former prevents the decoder from reconstructing the video background given video features, and thus helps reduce the background information in feature learning. The latter aims to confuse scene type classification given video features, with a specific emphasis on the action foreground, and helps to learn scene-invariant information. In addition, we design an experiment to quantify the scene bias. The results indicate that the current open-set action recognizers are biased toward the scene, and our proposed SOAR method better mitigates such bias. Furthermore, our extensive experiments demonstrate that our method outperforms state-of-the-art methods, and the ablation studies confirm the effectiveness of our proposed modules.

Yuli Zheng, Zhenyu Wu, Ye Yuan et al.

Credit risk modeling has permeated our everyday life. Most banks and financial companies use this technique to model their clients' trustworthiness. While machine learning is increasingly used in this field, the resulting large-scale collection of user private information has reinvigorated the privacy debate, considering dozens of data breach incidents every year caused by unauthorized hackers, and (potentially even more) information misuse/abuse by authorized parties. To address those critical concerns, this paper proposes a framework of Privacy-preserving Credit risk modeling based on Adversarial Learning (PCAL). PCAL aims to mask the private information inside the original dataset, while maintaining the important utility information for the target prediction task performance, by (iteratively) weighing between a privacy-risk loss and a utility-oriented loss. PCAL is compared against off-the-shelf options in terms of both utility and privacy protection. Results indicate that PCAL can learn an effective, privacy-free representation from user data, providing a solid foundation towards privacy-preserving machine learning for credit risk analysis.

Jianrong Wang, Ge Zhang, Zhenyu Wu et al.

In self-supervised monocular depth estimation, the depth discontinuity and motion objects' artifacts are still challenging problems. Existing self-supervised methods usually utilize a single view to train the depth estimation network. Compared with static views, abundant dynamic properties between video frames are beneficial to refined depth estimation, especially for dynamic objects. In this work, we propose a novel self-supervised joint learning framework for depth estimation using consecutive frames from monocular and stereo videos. The main idea is using an implicit depth cue extractor which leverages dynamic and static cues to generate useful depth proposals. These cues can predict distinguishable motion contours and geometric scene structures. Furthermore, a new high-dimensional attention module is introduced to extract clear global transformation, which effectively suppresses uncertainty of local descriptors in high-dimensional space, resulting in a more reliable optimization in learning framework. Experiments demonstrate that the proposed framework outperforms the state-of-the-art(SOTA) on KITTI and Make3D datasets.

Li-Phen Yen, Zhen-Yu Wu, Kuan-Yu Chen

Recent developments in deep learning have led to a significant innovation in various classic and practical subjects, including speech recognition, computer vision, question answering, information retrieval and so on. In the context of natural language processing (NLP), language representations have shown giant successes in many downstream tasks, so the school of studies have become a major stream of research recently. Because the immenseness of multimedia data along with speech have spread around the world in our daily life, spoken document retrieval (SDR) has become an important research subject in the past decades. Targeting on enhancing the SDR performance, the paper concentrates on proposing a neural retrieval framework, which assembles the merits of using language modeling (LM) mechanism in SDR and leveraging the abstractive information learned by the language representation models. Consequently, to our knowledge, this is a pioneer study on supervised training of a neural LM-based SDR framework, especially combined with the pretrained language representation methods.

Pritish Uplavikar, Zhenyu Wu, Zhangyang Wang

Raw underwater images are degraded due to wavelength dependent light attenuation and scattering, limiting their applicability in vision systems. Another factor that makes enhancing underwater images particularly challenging is the diversity of the water types in which they are captured. For example, images captured in deep oceanic waters have a different distribution from those captured in shallow coastal waters. Such diversity makes it hard to train a single model to enhance underwater images. In this work, we propose a novel model which nicely handles the diversity of water during the enhancement, by adversarially learning the content features of the images by disentangling the unwanted nuisances corresponding to water types (viewed as different domains). We use the learned domain agnostic features to generate enhanced underwater images. We train our model on a dataset consisting images of 10 Jerlov water types. Experimental results show that the proposed model not only outperforms the previous methods in SSIM and PSNR scores for almost all Jerlov water types but also generalizes well on real-world datasets. The performance of a high-level vision task (object detection) also shows improvement using enhanced images with our model.

Zhenyu Wu, Zhangyang Wang, Zhaowen Wang et al.

This paper aims to improve privacy-preserving visual recognition, an increasingly demanded feature in smart camera applications, by formulating a unique adversarial training framework. The proposed framework explicitly learns a degradation transform for the original video inputs, in order to optimize the trade-off between target task performance and the associated privacy budgets on the degraded video. A notable challenge is that the privacy budget, often defined and measured in task-driven contexts, cannot be reliably indicated using any single model performance, because a strong protection of privacy has to sustain against any possible model that tries to hack privacy information. Such an uncommon situation has motivated us to propose two strategies, i.e., budget model restarting and ensemble, to enhance the generalization of the learned degradation on protecting privacy against unseen hacker models. Novel training strategies, evaluation protocols, and result visualization methods have been designed accordingly. Two experiments on privacy-preserving action recognition, with privacy budgets defined in various ways, manifest the compelling effectiveness of the proposed framework in simultaneously maintaining high target task (action recognition) performance while suppressing the privacy breach risk.