Jian Li

Kaer Huang, Bingchuan Sun, Feng Chen et al.

In recent years, dominant Multi-object tracking (MOT) and segmentation (MOTS) methods mainly follow the tracking-by-detection paradigm. Transformer-based end-to-end (E2E) solutions bring some ideas to MOT and MOTS, but they cannot achieve a new state-of-the-art (SOTA) performance in major MOT and MOTS benchmarks. Detection and association are two main modules of the tracking-by-detection paradigm. Association techniques mainly depend on the combination of motion and appearance information. As deep learning has been recently developed, the performance of the detection and appearance model is rapidly improved. These trends made us consider whether we can achieve SOTA based on only high-performance detection and appearance model. Our paper mainly focuses on exploring this direction based on CBNetV2 with Swin-B as a detection model and MoCo-v2 as a self-supervised appearance model. Motion information and IoU mapping were removed during the association. Our method wins 1st place on the MOTS track and wins 2nd on the MOT track in the CVPR2023 WAD workshop. We hope our simple and effective method can give some insights to the MOT and MOTS research community. Source code will be released under this git repository

Simian Luo, Yiqin Tan, Longbo Huang et al.

Latent Diffusion models (LDMs) have achieved remarkable results in synthesizing high-resolution images. However, the iterative sampling process is computationally intensive and leads to slow generation. Inspired by Consistency Models (song et al.), we propose Latent Consistency Models (LCMs), enabling swift inference with minimal steps on any pre-trained LDMs, including Stable Diffusion (rombach et al). Viewing the guided reverse diffusion process as solving an augmented probability flow ODE (PF-ODE), LCMs are designed to directly predict the solution of such ODE in latent space, mitigating the need for numerous iterations and allowing rapid, high-fidelity sampling. Efficiently distilled from pre-trained classifier-free guided diffusion models, a high-quality 768 x 768 2~4-step LCM takes only 32 A100 GPU hours for training. Furthermore, we introduce Latent Consistency Fine-tuning (LCF), a novel method that is tailored for fine-tuning LCMs on customized image datasets. Evaluation on the LAION-5B-Aesthetics dataset demonstrates that LCMs achieve state-of-the-art text-to-image generation performance with few-step inference. Project Page: https://latent-consistency-models.github.io/

Gaurav Bagwe, Jian Li, Xiaoyong Yuan et al.

Despite the success of AI-enabled onboard perception, on-ramp merging has been one of the main challenges for autonomous driving. Due to limited sensing range of onboard sensors, a merging vehicle can hardly observe main road conditions and merge properly. By leveraging the wireless communications between connected and automated vehicles (CAVs), a merging CAV has potential to proactively obtain the intentions of nearby vehicles. However, CAVs can be prone to inaccurate observations, such as the noisy basic safety messages (BSM) and poor quality surveillance images. In this paper, we present a novel approach for Robust on-ramp merge of CAVs via Augmented and Multi-modal Reinforcement Learning, named by RAMRL. Specifically, we formulate the on-ramp merging problem as a Markov decision process (MDP) by taking driving safety, comfort driving behavior, and traffic efficiency into account. To provide reliable merging maneuvers, we simultaneously leverage BSM and surveillance images for multi-modal observation, which is used to learn a policy model through proximal policy optimization (PPO). Moreover, to improve data efficiency and provide better generalization performance, we train the policy model with augmented data (e.g., noisy BSM and noisy surveillance images). Extensive experiments are conducted with Simulation of Urban MObility (SUMO) platform under two typical merging scenarios. Experimental results demonstrate the effectiveness and efficiency of our robust on-ramp merging design.

Robert David, Jared Duke, Advait Jain et al.

Deep learning inference on embedded devices is a burgeoning field with myriad applications because tiny embedded devices are omnipresent. But we must overcome major challenges before we can benefit from this opportunity. Embedded processors are severely resource constrained. Their nearest mobile counterparts exhibit at least a 100 -- 1,000x difference in compute capability, memory availability, and power consumption. As a result, the machine-learning (ML) models and associated ML inference framework must not only execute efficiently but also operate in a few kilobytes of memory. Also, the embedded devices' ecosystem is heavily fragmented. To maximize efficiency, system vendors often omit many features that commonly appear in mainstream systems, including dynamic memory allocation and virtual memory, that allow for cross-platform interoperability. The hardware comes in many flavors (e.g., instruction-set architecture and FPU support, or lack thereof). We introduce TensorFlow Lite Micro (TF Micro), an open-source ML inference framework for running deep-learning models on embedded systems. TF Micro tackles the efficiency requirements imposed by embedded-system resource constraints and the fragmentation challenges that make cross-platform interoperability nearly impossible. The framework adopts a unique interpreter-based approach that provides flexibility while overcoming these challenges. This paper explains the design decisions behind TF Micro and describes its implementation details. Also, we present an evaluation to demonstrate its low resource requirement and minimal run-time performance overhead.

Zhen He, Jian Li, Daxue Liu et al.

Online Multi-Object Tracking (MOT) from videos is a challenging computer vision task which has been extensively studied for decades. Most of the existing MOT algorithms are based on the Tracking-by-Detection (TBD) paradigm combined with popular machine learning approaches which largely reduce the human effort to tune algorithm parameters. However, the commonly used supervised learning approaches require the labeled data (e.g., bounding boxes), which is expensive for videos. Also, the TBD framework is usually suboptimal since it is not end-to-end, i.e., it considers the task as detection and tracking, but not jointly. To achieve both label-free and end-to-end learning of MOT, we propose a Tracking-by-Animation framework, where a differentiable neural model first tracks objects from input frames and then animates these objects into reconstructed frames. Learning is then driven by the reconstruction error through backpropagation. We further propose a Reprioritized Attentive Tracking to improve the robustness of data association. Experiments conducted on both synthetic and real video datasets show the potential of the proposed model. Our project page is publicly available at: https://github.com/zhen-he/tracking-by-animation

Caleb S. Spradlin, Jordan A. Caraballo-Vega, Jian Li et al.

Foundation models have the potential to transform the landscape of remote sensing (RS) data analysis by enabling large computer vision models to be pre-trained on vast amounts of remote sensing data. These models can then be fine-tuned with small amounts of labeled training and applied to a variety of applications. Most existing foundation models are designed for high spatial resolution, cloud-free satellite imagery or photos, limiting their applicability in scenarios that require frequent temporal monitoring or broad spectral profiles. As a result, foundation models trained solely on cloud-free images have limited utility for applications that involve atmospheric variables or require atmospheric corrections. We introduce SatVision-TOA, a novel foundation model pre-trained on 14-band MODIS L1B Top-Of-Atmosphere (TOA) radiance imagery, addressing the need for models pre-trained to handle moderate- and coarse-resolution all-sky remote sensing data. The SatVision-TOA model is pre-trained using a Masked-Image-Modeling (MIM) framework and the SwinV2 architecture, and learns detailed contextual representations through self-supervised learning without the need for labels. It is a 3 billion parameter model that is trained on 100 million images. To our knowledge this is the largest foundation model trained solely on satellite RS imagery. Results show that SatVision-TOA achieves superior performance over baseline methods on downstream tasks such as 3D cloud retrieval. Notably, the model achieves a mean intersection over union (mIOU) of 0.46, a substantial improvement over the baseline mIOU of 0.22. Additionally, the rate of false negative results in the fine-tuning task were reduced by over 50% compared to the baseline. Our work advances pre-trained vision modeling for multispectral RS by learning from a variety of atmospheric and aerosol conditions to improve cloud and land surface monitoring.

Zihao Wang, Wei Peng, Junming Zhang et al.

Encrypted traffic classification aims to identify applications or services by analyzing network traffic data. One of the critical challenges is the continuous emergence of new applications, which generates Out-of-Distribution (OOD) traffic patterns that deviate from known categories and are not well represented by predefined models. Current approaches rely on predefined categories, which limits their effectiveness in handling unknown traffic types. Although some methods mitigate this limitation by simply classifying unknown traffic into a single "Other" category, they fail to make a fine-grained classification. In this paper, we propose a Two-stage Adaptive OOD classification Network (TAO-Net) that achieves accurate classification for both In-Distribution (ID) and OOD encrypted traffic. The method incorporates an innovative two-stage design: the first stage employs a hybrid OOD detection mechanism that integrates transformer-based inter-layer transformation smoothness and feature analysis to effectively distinguish between ID and OOD traffic, while the second stage leverages large language models with a novel semantic-enhanced prompt strategy to transform OOD traffic classification into a generation task, enabling flexible fine-grained classification without relying on predefined labels. Experiments on three datasets demonstrate that TAO-Net achieves 96.81-97.70% macro-precision and 96.77-97.68% macro-F1, outperforming previous methods that only reach 44.73-86.30% macro-precision, particularly in identifying emerging network applications.

Jinxiang Lai, Wenlong Wu, Jiawei Zhan et al.

Box-supervised instance segmentation methods aim to achieve instance segmentation with only box annotations. Recent methods have demonstrated the effectiveness of acquiring high-quality pseudo masks under the teacher-student framework. Building upon this foundation, we propose a BoxSeg framework involving two novel and general modules named the Quality-Aware Module (QAM) and the Peer-assisted Copy-paste (PC). The QAM obtains high-quality pseudo masks and better measures the mask quality to help reduce the effect of noisy masks, by leveraging the quality-aware multi-mask complementation mechanism. The PC imitates Peer-Assisted Learning to further improve the quality of the low-quality masks with the guidance of the obtained high-quality pseudo masks. Theoretical and experimental analyses demonstrate the proposed QAM and PC are effective. Extensive experimental results show the superiority of our BoxSeg over the state-of-the-art methods, and illustrate the QAM and PC can be applied to improve other models.

Yi Sun, Xin Xu, Jian Li et al.

Multi-output deep neural networks(MONs) contain multiple task branches, and these tasks usually share partial network filters that lead to the entanglement of different task inference routes. Due to the inconsistent optimization objectives, the task gradients used for training MONs will interfere with each other on the shared routes, which will decrease the overall model performance. To address this issue, we propose a novel gradient de-conflict algorithm named DR-MGF(Dynamic Routes and Meta-weighted Gradient Fusion) in this work. Different from existing de-conflict methods, DR-MGF achieves gradient de-conflict in MONs by learning task-preferred inference routes. The proposed method is motivated by our experimental findings: the shared filters are not equally important to different tasks. By designing the learnable task-specific importance variables, DR-MGF evaluates the importance of filters for different tasks. Through making the dominances of tasks over filters be proportional to the task-specific importance of filters, DR-MGF can effectively reduce the inter-task interference. The task-specific importance variables ultimately determine task-preferred inference routes at the end of training iterations. Extensive experimental results on CIFAR, ImageNet, and NYUv2 illustrate that DR-MGF outperforms the existing de-conflict methods both in prediction accuracy and convergence speed of MONs. Furthermore, DR-MGF can be extended to general MONs without modifying the overall network structures.

Jian Li, Bin Zhang, Yabiao Wang et al.

Along with current multi-scale based detectors, Feature Aggregation and Enhancement (FAE) modules have shown superior performance gains for cutting-edge object detection. However, these hand-crafted FAE modules show inconsistent improvements on face detection, which is mainly due to the significant distribution difference between its training and applying corpus, COCO vs. WIDER Face. To tackle this problem, we essentially analyse the effect of data distribution, and consequently propose to search an effective FAE architecture, termed AutoFAE by a differentiable architecture search, which outperforms all existing FAE modules in face detection with a considerable margin. Upon the found AutoFAE and existing backbones, a supernet is further built and trained, which automatically obtains a family of detectors under the different complexity constraints. Extensive experiments conducted on popular benchmarks, WIDER Face and FDDB, demonstrate the state-of-the-art performance-efficiency trade-off for the proposed automatic and scalable face detector (ASFD) family. In particular, our strong ASFD-D6 outperforms the best competitor with AP 96.7/96.2/92.1 on WIDER Face test, and the lightweight ASFD-D0 costs about 3.1 ms, more than 320 FPS, on the V100 GPU with VGA-resolution images.

Jian Li, Raziel Alvarez

Integer quantization of neural networks can be defined as the approximation of the high precision computation of the canonical neural network formulation, using reduced integer precision. It plays a significant role in the efficient deployment and execution of machine learning (ML) systems, reducing memory consumption and leveraging typically faster computations. In this work, we present an integer-only quantization strategy for Long Short-Term Memory (LSTM) neural network topologies, which themselves are the foundation of many production ML systems. Our quantization strategy is accurate (e.g. works well with quantization post-training), efficient and fast to execute (utilizing 8 bit integer weights and mostly 8 bit activations), and is able to target a variety of hardware (by leveraging instructions sets available in common CPU architectures, as well as available neural accelerators).

Bin Zhang, Jian Li, Yabiao Wang et al.

Cartoon face detection is a more challenging task than human face detection due to many difficult scenarios is involved. Aiming at the characteristics of cartoon faces, such as huge differences within the intra-faces, in this paper, we propose an asymmetric cartoon face detector, named ACFD. Specifically, it consists of the following modules: a novel backbone VoVNetV3 comprised of several asymmetric one-shot aggregation modules (AOSA), asymmetric bi-directional feature pyramid network (ABi-FPN), dynamic anchor match strategy (DAM) and the corresponding margin binary classification loss (MBC). In particular, to generate features with diverse receptive fields, multi-scale pyramid features are extracted by VoVNetV3, and then fused and enhanced simultaneously by ABi-FPN for handling the faces in some extreme poses and have disparate aspect ratios. Besides, DAM is used to match enough high-quality anchors for each face, and MBC is for the strong power of discrimination. With the effectiveness of these modules, our ACFD achieves the 1st place on the detection track of 2020 iCartoon Face Challenge under the constraints of model size 200MB, inference time 50ms per image, and without any pretrained models.

Jian Li, Yong Liu, Jiankun Liu et al.

Due to their high computational efficiency on a continuous space, gradient optimization methods have shown great potential in the neural architecture search (NAS) domain. The mapping of network representation from the discrete space to a latent space is the key to discovering novel architectures, however, existing gradient-based methods fail to fully characterize the networks. In this paper, we propose an efficient NAS approach to optimize network architectures in a continuous space, where the latent space is built upon variational autoencoder (VAE) and graph neural networks (GNN). The framework jointly learns four components: the encoder, the performance predictor, the complexity predictor and the decoder in an end-to-end manner. The encoder and the decoder belong to a graph VAE, mapping architectures between continuous representations and network architectures. The predictors are two regression models, fitting the performance and computational complexity, respectively. Those predictors ensure the discovered architectures characterize both excellent performance and high computational efficiency. Extensive experiments demonstrate our framework not only generates appropriate continuous representations but also discovers powerful neural architectures.

Bin Zhang, Jian Li, Yabiao Wang et al.

In this paper, we propose a novel Automatic and Scalable Face Detector (ASFD), which is based on a combination of neural architecture search techniques as well as a new loss design. First, we propose an automatic feature enhance module named Auto-FEM by improved differential architecture search, which allows efficient multi-scale feature fusion and context enhancement. Second, we use Distance-based Regression and Margin-based Classification (DRMC) multi-task loss to predict accurate bounding boxes and learn highly discriminative deep features. Third, we adopt compound scaling methods and uniformly scale the backbone, feature modules, and head networks to develop a family of ASFD, which are consistently more efficient than the state-of-the-art face detectors. Extensive experiments conducted on popular benchmarks, e.g. WIDER FACE and FDDB, demonstrate that our ASFD-D6 outperforms the prior strong competitors, and our lightweight ASFD-D0 runs at more than 120 FPS with Mobilenet for VGA-resolution images.

Chuming Lin, Jian Li, Yabiao Wang et al.

Generating temporal action proposals remains a very challenging problem, where the main issue lies in predicting precise temporal proposal boundaries and reliable action confidence in long and untrimmed real-world videos. In this paper, we propose an efficient and unified framework to generate temporal action proposals named Dense Boundary Generator (DBG), which draws inspiration from boundary-sensitive methods and implements boundary classification and action completeness regression for densely distributed proposals. In particular, the DBG consists of two modules: Temporal boundary classification (TBC) and Action-aware completeness regression (ACR). The TBC aims to provide two temporal boundary confidence maps by low-level two-stream features, while the ACR is designed to generate an action completeness score map by high-level action-aware features. Moreover, we introduce a dual stream BaseNet (DSB) to encode RGB and optical flow information, which helps to capture discriminative boundary and actionness features. Extensive experiments on popular benchmarks ActivityNet-1.3 and THUMOS14 demonstrate the superiority of DBG over the state-of-the-art proposal generator (e.g., MGG and BMN). Our code will be made available upon publication.

Jie Tang, Fei-Peng Tian, Wei Feng et al.

Dense depth perception is critical for autonomous driving and other robotics applications. However, modern LiDAR sensors only provide sparse depth measurement. It is thus necessary to complete the sparse LiDAR data, where a synchronized guidance RGB image is often used to facilitate this completion. Many neural networks have been designed for this task. However, they often naïvely fuse the LiDAR data and RGB image information by performing feature concatenation or element-wise addition. Inspired by the guided image filtering, we design a novel guided network to predict kernel weights from the guidance image. These predicted kernels are then applied to extract the depth image features. In this way, our network generates content-dependent and spatially-variant kernels for multi-modal feature fusion. Dynamically generated spatially-variant kernels could lead to prohibitive GPU memory consumption and computation overhead. We further design a convolution factorization to reduce computation and memory consumption. The GPU memory reduction makes it possible for feature fusion to work in multi-stage scheme. We conduct comprehensive experiments to verify our method on real-world outdoor, indoor and synthetic datasets. Our method produces strong results. It outperforms state-of-the-art methods on the NYUv2 dataset and ranks 1st on the KITTI depth completion benchmark at the time of submission. It also presents strong generalization capability under different 3D point densities, various lighting and weather conditions as well as cross-dataset evaluations. The code will be released for reproduction.

Jian Li, Yabiao Wang, Changan Wang et al.

In this paper, we propose a novel face detection network with three novel contributions that address three key aspects of face detection, including better feature learning, progressive loss design and anchor assign based data augmentation, respectively. First, we propose a Feature Enhance Module (FEM) for enhancing the original feature maps to extend the single shot detector to dual shot detector. Second, we adopt Progressive Anchor Loss (PAL) computed by two different sets of anchors to effectively facilitate the features. Third, we use an Improved Anchor Matching (IAM) by integrating novel anchor assign strategy into data augmentation to provide better initialization for the regressor. Since these techniques are all related to the two-stream design, we name the proposed network as Dual Shot Face Detector (DSFD). Extensive experiments on popular benchmarks, WIDER FACE and FDDB, demonstrate the superiority of DSFD over the state-of-the-art face detectors.

Jian Li

Of later years, numerous bottom-up attention models have been proposed on different assumptions. However, the produced saliency maps may be different from each other even from the same input image. We also observe that human fixation map varies across time greatly. When people freely view an image, they tend to allocate attention at salient regions of large scale at first, and then search more and more detailed regions. In this paper, we argue that, for one input image visual attention cannot be described by only one single saliency map, and this mechanism should be modeled as a dynamic process. Under the frequency domain paradigm, we proposed a global inhibition model to mimic this process by suppressing the {\it non-saliency} in the input image; we also show that the dynamic process is influenced by one parameter in the frequency domain. Experiments illustrate that the proposed model is capable of predicting human dynamic fixation distribution.

Jianbo Guo, Guangxiang Zhu, Jian Li

Generative Adversarial Networks (GANs) have shown impressive performance in generating photo-realistic images. They fit generative models by minimizing certain distance measure between the real image distribution and the generated data distribution. Several distance measures have been used, such as Jensen-Shannon divergence, $f$-divergence, and Wasserstein distance, and choosing an appropriate distance measure is very important for training the generative network. In this paper, we choose to use the maximum mean discrepancy (MMD) as the distance metric, which has several nice theoretical guarantees. In fact, generative moment matching network (GMMN) (Li, Swersky, and Zemel 2015) is such a generative model which contains only one generator network $G$ trained by directly minimizing MMD between the real and generated distributions. However, it fails to generate meaningful samples on challenging benchmark datasets, such as CIFAR-10 and LSUN. To improve on GMMN, we propose to add an extra network $F$, called mapper. $F$ maps both real data distribution and generated data distribution from the original data space to a feature representation space $\mathcal{R}$, and it is trained to maximize MMD between the two mapped distributions in $\mathcal{R}$, while the generator $G$ tries to minimize the MMD. We call the new model generative adversarial mapping networks (GAMNs). We demonstrate that the adversarial mapper $F$ can help $G$ to better capture the underlying data distribution. We also show that GAMN significantly outperforms GMMN, and is also superior to or comparable with other state-of-the-art GAN based methods on MNIST, CIFAR-10 and LSUN-Bedrooms datasets.

Keyu Lu, Jian Li, Xiangjing An et al.

Vision-based object detection is one of the fundamental functions in numerous traffic scene applications such as self-driving vehicle systems and advance driver assistance systems (ADAS). However, it is also a challenging task due to the diversity of traffic scene and the storage, power and computing source limitations of the platforms for traffic scene applications. This paper presents a generalized Haar filter based deep network which is suitable for the object detection tasks in traffic scene. In this approach, we first decompose a object detection task into several easier local regression tasks. Then, we handle the local regression tasks by using several tiny deep networks which simultaneously output the bounding boxes, categories and confidence scores of detected objects. To reduce the consumption of storage and computing resources, the weights of the deep networks are constrained to the form of generalized Haar filter in training phase. Additionally, we introduce the strategy of sparse windows generation to improve the efficiency of the algorithm. Finally, we perform several experiments to validate the performance of our proposed approach. Experimental results demonstrate that the proposed approach is both efficient and effective in traffic scene compared with the state-of-the-art.

Weishan Dong, Jian Li, Renjie Yao et al.

Characterizing driving styles of human drivers using vehicle sensor data, e.g., GPS, is an interesting research problem and an important real-world requirement from automotive industries. A good representation of driving features can be highly valuable for autonomous driving, auto insurance, and many other application scenarios. However, traditional methods mainly rely on handcrafted features, which limit machine learning algorithms to achieve a better performance. In this paper, we propose a novel deep learning solution to this problem, which could be the first attempt of extending deep learning to driving behavior analysis based on GPS data. The proposed approach can effectively extract high level and interpretable features describing complex driving patterns. It also requires significantly less human experience and work. The power of the learned driving style representations are validated through the driver identification problem using a large real dataset.

Jian Li, Martin Levine, Xiangjing An et al.

We address the issue of visual saliency from three perspectives. First, we consider saliency detection as a frequency domain analysis problem. Second, we achieve this by employing the concept of {\it non-saliency}. Third, we simultaneously consider the detection of salient regions of different size. The paper proposes a new bottom-up paradigm for detecting visual saliency, characterized by a scale-space analysis of the amplitude spectrum of natural images. We show that the convolution of the {\it image amplitude spectrum} with a low-pass Gaussian kernel of an appropriate scale is equivalent to such an image saliency detector. The saliency map is obtained by reconstructing the 2-D signal using the original phase and the amplitude spectrum, filtered at a scale selected by minimizing saliency map entropy. A Hypercomplex Fourier Transform performs the analysis in the frequency domain. Using available databases, we demonstrate experimentally that the proposed model can predict human fixation data. We also introduce a new image database and use it to show that the saliency detector can highlight both small and large salient regions, as well as inhibit repeated distractors in cluttered images. In addition, we show that it is able to predict salient regions on which people focus their attention.

Shusen Wang, Zhihua Zhang, Jian Li

The CUR matrix decomposition is an important extension of Nyström approximation to a general matrix. It approximates any data matrix in terms of a small number of its columns and rows. In this paper we propose a novel randomized CUR algorithm with an expected relative-error bound. The proposed algorithm has the advantages over the existing relative-error CUR algorithms that it possesses tighter theoretical bound and lower time complexity, and that it can avoid maintaining the whole data matrix in main memory. Finally, experiments on several real-world datasets demonstrate significant improvement over the existing relative-error algorithms.