Pan He

Xinyu Liu, Darryl Cherian Jacob, Yang Zhou et al.

Recent reinforcement learning (RL) post-training approaches primarily optimize the final output policy using sparse outcome-level rewards, while largely overlooking predictive signals encoded in intermediate representations. In this paper, we introduce a new paradigm called on-policy internal self-distillation and propose the OISD framework, which improves reasoning by transferring on-policy predictive signals from the final layer to intermediate representations. During rollout and Group Relative Policy Optimization (GRPO) optimization, the final layer acts as both the policy and a detached internal teacher for selected intermediate layers, which are guided to align with it through two complementary mechanisms: logit alignment, which transfers high-level reasoning behaviors (how to think), and attention alignment, which enforces consistent attention patterns (where to look) from the final layer to the selected intermediate layer, both without requiring external privileged information. Our OISD, together with GRPO, employs signed advantage-weighted Jensen--Shannon alignment to distill informative intermediate representations while preserving policy consistency under a unified acting policy. Experimental results demonstrate the effectiveness of OISD, with substantial and consistent improvements over strong reasoning RL baselines across four mathematical reasoning tasks. The code will be released at https://github.com/THE-MALT-LAB/OISD

Anand Rangarajan, Pan He, Jaemoon Lee et al.

We seek to impose linear, equality constraints in feedforward neural networks. As top layer predictors are usually nonlinear, this is a difficult task if we seek to deploy standard convex optimization methods and strong duality. To overcome this, we introduce a new saddle-point Lagrangian with auxiliary predictor variables on which constraints are imposed. Elimination of the auxiliary variables leads to a dual minimization problem on the Lagrange multipliers introduced to satisfy the linear constraints. This minimization problem is combined with the standard learning problem on the weight matrices. From this theoretical line of development, we obtain the surprising interpretation of Lagrange parameters as additional, penultimate layer hidden units with fixed weights stemming from the constraints. Consequently, standard minimization approaches can be used despite the inclusion of Lagrange parameters -- a very satisfying, albeit unexpected, discovery. Examples ranging from multi-label classification to constrained autoencoders are envisaged in the future. The code has been made available at https://github.com/anandrajan0/smartalec

Pan He, Patrick Emami, Sanjay Ranka et al.

In this paper, we present a new self-supervised scene flow estimation approach for a pair of consecutive point clouds. The key idea of our approach is to represent discrete point clouds as continuous probability density functions using Gaussian mixture models. Scene flow estimation is therefore converted into the problem of recovering motion from the alignment of probability density functions, which we achieve using a closed-form expression of the classic Cauchy-Schwarz divergence. Unlike existing nearest-neighbor-based approaches that use hard pairwise correspondences, our proposed approach establishes soft and implicit point correspondences between point clouds and generates more robust and accurate scene flow in the presence of missing correspondences and outliers. Comprehensive experiments show that our method makes noticeable gains over the Chamfer Distance and the Earth Mover's Distance in real-world environments and achieves state-of-the-art performance among self-supervised learning methods on FlyingThings3D and KITTI, even outperforming some supervised methods with ground truth annotations.

Yinyi Luo, Wenwen Wang, Hayes Bai et al.

Recent advances in unified multimodal models (UMMs) have led to a proliferation of architectures capable of understanding, generating, and editing across visual and textual modalities. However, developing a unified framework for UMMs remains challenging due to the diversity of model architectures and the heterogeneity of training paradigms and implementation details. In this paper, we present TorchUMM, the first unified codebase for comprehensive evaluation, analysis, and post-training across diverse UMM backbones, tasks, and datasets. TorchUMM supports a broad spectrum of models covering a wide range of scales and design paradigms. Our benchmark encompasses three core task dimensions: multimodal understanding, generation, and editing, and integrates both established and novel datasets to evaluate perception, reasoning, compositionality, and instruction-following abilities. By providing a unified interface and standardized evaluation protocols, TorchUMM enables fair and reproducible comparisons across heterogeneous models and fosters deeper insights into their strengths and limitations, facilitating the development of more capable unified multimodal systems. Code is available at: https://github.com/AIFrontierLab/TorchUMM.

Darryl Cherian Jacob, Xinyu Liu, Kai Wang et al.

Vision-language models (VLMs) have shown strong performance in video anomaly detection (VAD) while providing interpretable predictions. However, existing VLM-based VAD methods suffer from a fundamental mismatch between training and inference in both data distribution and model configuration. First, most approaches rely on static post-training adaptation, limiting generalization under distribution shifts such as unseen environments or anomaly types. Second, they train VLMs on sparse frames from long videos, but perform inference on densely sampled short segments, creating inconsistencies between training and testing. To address these limitations, we propose COPRA, a conditional parameter adaptation framework for VLM-based VAD. Instead of fixed prompts or shared parameter updates, COPRA generates input-specific parameter updates to dynamically adapt a frozen VLM for each video segment during both training and inference. Experiments show strong performance on standard VAD benchmarks, consistently outperforming static baselines in both in-domain and cross-domain settings. Moreover, COPRA generalizes beyond VAD to unseen tasks such as multiple-choice Video Question Answering and Dense Captioning. These results highlight COPRA as an effective weight-space generation framework for scalable, adaptive, and context-aware video understanding. The code will be released at https://github.com/THE-MALT-LAB/COPRA

Aotian Wu, Pan He, Xiao Li et al.

Most existing perception systems rely on sensory data acquired from cameras, which perform poorly in low light and adverse weather conditions. To resolve this limitation, we have witnessed advanced LiDAR sensors become popular in perception tasks in autonomous driving applications. Nevertheless, their usage in traffic monitoring systems is less ubiquitous. We identify two significant obstacles in cost-effectively and efficiently developing such a LiDAR-based traffic monitoring system: (i) public LiDAR datasets are insufficient for supporting perception tasks in infrastructure systems, and (ii) 3D annotations on LiDAR point clouds are time-consuming and expensive. To fill this gap, we present an efficient semi-automated annotation tool that automatically annotates LiDAR sequences with tracking algorithms while offering a fully annotated infrastructure LiDAR dataset -- FLORIDA (Florida LiDAR-based Object Recognition and Intelligent Data Annotation) -- which will be made publicly available. Our advanced annotation tool seamlessly integrates multi-object tracking (MOT), single-object tracking (SOT), and suitable trajectory post-processing techniques. Specifically, we introduce a human-in-the-loop schema in which annotators recursively fix and refine annotations imperfectly predicted by our tool and incrementally add them to the training dataset to obtain better SOT and MOT models. By repeating the process, we significantly increase the overall annotation speed by three to four times and obtain better qualitative annotations than a state-of-the-art annotation tool. The human annotation experiments verify the effectiveness of our annotation tool. In addition, we provide detailed statistics and object detection evaluation results for our dataset in serving as a benchmark for perception tasks at traffic intersections.

Xiao Li, Pan He, Aotian Wu et al.

We address the problem of unsupervised semantic segmentation of outdoor LiDAR point clouds in diverse traffic scenarios. The key idea is to leverage the spatiotemporal nature of a dynamic point cloud sequence and introduce drastically stronger augmentation by establishing spatiotemporal correspondences across multiple frames. We dovetail clustering and pseudo-label learning in this work. Essentially, we alternate between clustering points into semantic groups and optimizing models using point-wise pseudo-spatiotemporal labels with a simple learning objective. Therefore, our method can learn discriminative features in an unsupervised learning fashion. We show promising segmentation performance on Semantic-KITTI, SemanticPOSS, and FLORIDA benchmark datasets covering scenarios in autonomous vehicle and intersection infrastructure, which is competitive when compared against many existing fully supervised learning methods. This general framework can lead to a unified representation learning approach for LiDAR point clouds incorporating domain knowledge.

Pan He, Patrick Emami, Sanjay Ranka et al.

We present a new approach to unsupervised shape correspondence learning between pairs of point clouds. We make the first attempt to adapt the classical locally linear embedding algorithm (LLE) -- originally designed for nonlinear dimensionality reduction -- for shape correspondence. The key idea is to find dense correspondences between shapes by first obtaining high-dimensional neighborhood-preserving embeddings of low-dimensional point clouds and subsequently aligning the source and target embeddings using locally linear transformations. We demonstrate that learning the embedding using a new LLE-inspired point cloud reconstruction objective results in accurate shape correspondences. More specifically, the approach comprises an end-to-end learnable framework of extracting high-dimensional neighborhood-preserving embeddings, estimating locally linear transformations in the embedding space, and reconstructing shapes via divergence measure-based alignment of probabilistic density functions built over reconstructed and target shapes. Our approach enforces embeddings of shapes in correspondence to lie in the same universal/canonical embedding space, which eventually helps regularize the learning process and leads to a simple nearest neighbors approach between shape embeddings for finding reliable correspondences. Comprehensive experiments show that the new method makes noticeable improvements over state-of-the-art approaches on standard shape correspondence benchmark datasets covering both human and nonhuman shapes.

Patrick Emami, Pan He, Sanjay Ranka et al.

Unconditional scene inference and generation are challenging to learn jointly with a single compositional model. Despite encouraging progress on models that extract object-centric representations (''slots'') from images, unconditional generation of scenes from slots has received less attention. This is primarily because learning the multi-object relations necessary to imagine coherent scenes is difficult. We hypothesize that most existing slot-based models have a limited ability to learn object correlations. We propose two improvements that strengthen object correlation learning. The first is to condition the slots on a global, scene-level variable that captures higher-order correlations between slots. Second, we address the fundamental lack of a canonical order for objects in images by proposing to learn a consistent order to use for the autoregressive generation of scene objects. Specifically, we train an autoregressive slot prior to sequentially generate scene objects following a learned order. Ordered slot inference entails first estimating a randomly ordered set of slots using existing approaches for extracting slots from images, then aligning those slots to ordered slots generated autoregressively with the slot prior. Our experiments across three multi-object environments demonstrate clear gains in unconditional scene generation quality. Detailed ablation studies are also provided that validate the two proposed improvements.

Yuang Geng, Junkai Zhou, Kang Yang et al.

In this paper, we address the challenging problem of single-scene, fully unsupervised video anomaly detection (VAD), where raw videos containing both normal and abnormal events are used directly for training and testing without any labels. This differs sharply from prior work that either requires extensive labeling (fully or weakly supervised) or depends on normal-only videos (one-class classification), which are vulnerable to distribution shifts and contamination. We propose an entropy-guided autoencoder that detects anomalies through reconstruction error by reconstructing normal frames well while making anomalies reconstruct poorly. The key idea is to combine the standard reconstruction loss with a novel Minimal Latent Entropy (MLE) loss in the autoencoder. Reconstruction loss alone maps normal and abnormal inputs to distinct latent clusters due to their inherent differences, but also risks reconstructing anomalies too well to detect. Therefore, MLE loss addresses this by minimizing the entropy of latent embeddings, encouraging them to concentrate around high-density regions. Since normal frames dominate the raw video, sparse anomalous embeddings are pulled into the normal cluster, so the decoder emphasizes normal patterns and produces poor reconstructions for anomalies. This dual-loss design produces a clear reconstruction gap that enables effective anomaly detection. Extensive experiments on two widely used benchmarks and a challenging self-collected driving dataset demonstrate that our method achieves robust and superior performance over baselines.

Saket S. Chaturvedi, Gaurav Bagwe, Lan Zhang et al.

LiDAR-based 3D object detection is widely used in safety-critical systems. However, these systems remain vulnerable to backdoor attacks that embed hidden malicious behaviors during training. A key limitation of existing backdoor attacks is their lack of physical realizability, primarily due to the digital-to-physical domain gap. Digital triggers often fail in real-world settings because they overlook material-dependent LiDAR reflection properties. On the other hand, physically constructed triggers are often unoptimized, leading to low effectiveness or easy detectability.This paper introduces Material-Oriented Backdoor Attack (MOBA), a novel framework that bridges the digital-physical gap by explicitly modeling the material properties of real-world triggers. MOBA tackles two key challenges in physical backdoor design: 1) robustness of the trigger material under diverse environmental conditions, 2) alignment between the physical trigger's behavior and its digital simulation. First, we propose a systematic approach to selecting robust trigger materials, identifying titanium dioxide (TiO_2) for its high diffuse reflectivity and environmental resilience. Second, to ensure the digital trigger accurately mimics the physical behavior of the material-based trigger, we develop a novel simulation pipeline that features: (1) an angle-independent approximation of the Oren-Nayar BRDF model to generate realistic LiDAR intensities, and (2) a distance-aware scaling mechanism to maintain spatial consistency across varying depths. We conduct extensive experiments on state-of-the-art LiDAR-based and Camera-LiDAR fusion models, showing that MOBA achieves a 93.50% attack success rate, outperforming prior methods by over 41%. Our work reveals a new class of physically realizable threats and underscores the urgent need for defenses that account for material-level properties in real-world environments.

Darryl Jacob, Xinyu Liu, Muchao Ye et al.

Transparent decision-making is essential for traffic signal control (TSC) systems to earn public trust. However, traditional reinforcement learning-based TSC methods function as black boxes with limited interpretability. Although large language models (LLMs) can provide natural language reasoning, reinforcement finetuning for TSC remains unstable because feedback is sparse and delayed, while most actions produce only marginal changes in congestion metrics. We introduce OracleTSC, which stabilizes LLM-based TSC through two mechanisms: (1) a reward hurdle mechanism that filters weak learning signals by subtracting a calibrated threshold from environmental rewards, and (2) uncertainty regularization that maximizes the probability of the selected response to encourage consistent decisions across sampled outputs. Experiments on the LibSignal benchmark show that OracleTSC enables a compact LLaMA3-8B model to substantially improve traffic efficiency, achieving a 75% reduction in travel time and a 67% decrease in queue length compared with the pretrained baseline while preserving interpretability through natural language explanations. OracleTSC also demonstrates strong cross-intersection generalization: a policy trained on one intersection transfers to a structurally different intersection with 17% lower travel time and 39% lower queue length without additional finetuning. These results suggest that uncertainty-aware reward shaping can improve the stability and effectiveness of reinforcement fine-tuning for TSC.

Xinyu Liu, Darryl C. Jacob, Yuxin Liu et al.

Although recent traffic benchmarks have advanced multimodal data analysis, they generally lack systematic evaluation aligned with official safety standards. To fill this gap, we introduce RoadSafe365, a large-scale vision-language benchmark that supports fine-grained analysis of traffic safety from extensive and diverse real-world video data collections. Unlike prior works that focus primarily on coarse accident identification, RoadSafe365 is independently curated and systematically organized using a hierarchical taxonomy that refines and extends foundational definitions of crash, incident, and violation to bridge official traffic safety standards with data-driven traffic understanding systems. RoadSafe365 provides rich attribute annotations across diverse traffic event types, environmental contexts, and interaction scenarios, yielding 36,196 annotated clips from both dashcam and surveillance cameras. Each clip is paired with multiple-choice question-answer sets, comprising 864K candidate options, 8.4K unique answers, and 36K detailed scene descriptions collectively designed for vision-language understanding and reasoning. We establish strong baselines and observe consistent gains when fine-tuning on RoadSafe365. Cross-domain experiments on both real and synthetic datasets further validate its effectiveness. Designed for large-scale training and standardized evaluation, RoadSafe365 provides a comprehensive benchmark to advance reproducible research in real-world traffic safety analysis.

Mitchell Piehl, Zhaohan Xi, Zuobin Xiong et al.

Large language models (LLMs) are increasingly augmented with long-term memory systems to overcome finite context windows and enable persistent reasoning across interactions. However, recent research finds that LLMs become more vulnerable because memory provides extra attack surfaces. In this paper, we present the first systematic study of black-box adversarial memory injection attacks that target the similarity-based retrieval mechanism in long-term memory-augmented LLMs. We introduce ER-MIA, a unified framework that exposes this vulnerability and formalizes two realistic attack settings: content-based attacks and question-targeted attacks. In these settings, ER-MIA includes an arsenal of composable attack primitives and ensemble attacks that achieve high success rates under minimal attacker assumptions. Extensive experiments across multiple LLMs and long-term memory systems demonstrate that similarity-based retrieval constitutes a fundamental and system-level vulnerability, revealing security risks that persist across memory designs and application scenarios.

Muchao Ye, Weiyang Liu, Pan He

The rapid advancement of vision-language models (VLMs) has established a new paradigm in video anomaly detection (VAD): leveraging VLMs to simultaneously detect anomalies and provide comprehendible explanations for the decisions. Existing work in this direction often assumes the complex reasoning required for VAD exceeds the capabilities of pretrained VLMs. Consequently, these approaches either incorporate specialized reasoning modules during inference or rely on instruction tuning datasets through additional training to adapt VLMs for VAD. However, such strategies often incur substantial computational costs or data annotation overhead. To address these challenges in explainable VAD, we introduce a verbalized learning framework named VERA that enables VLMs to perform VAD without model parameter modifications. Specifically, VERA automatically decomposes the complex reasoning required for VAD into reflections on simpler, more focused guiding questions capturing distinct abnormal patterns. It treats these reflective questions as learnable parameters and optimizes them through data-driven verbal interactions between learner and optimizer VLMs, using coarsely labeled training data. During inference, VERA embeds the learned questions into model prompts to guide VLMs in generating segment-level anomaly scores, which are then refined into frame-level scores via the fusion of scene and temporal contexts. Experimental results on challenging benchmarks demonstrate that the learned questions of VERA are highly adaptable, significantly improving both detection performance and explainability of VLMs for VAD.

Pan He, Quanyi Li, Xiaoyong Yuan et al.

Traffic signal control (TSC) is crucial for reducing traffic congestion that leads to smoother traffic flow, reduced idling time, and mitigated CO2 emissions. In this study, we explore the computer vision approach for TSC that modulates on-road traffic flows through visual observation. Unlike traditional feature-based approaches, vision-based methods depend much less on heuristics and predefined features, bringing promising potentials for end-to-end learning and optimization of traffic signals. Thus, we introduce a holistic traffic simulation framework called TrafficDojo towards vision-based TSC and its benchmarking by integrating the microscopic traffic flow provided in SUMO into the driving simulator MetaDrive. This proposed framework offers a versatile traffic environment for in-depth analysis and comprehensive evaluation of traffic signal controllers across diverse traffic conditions and scenarios. We establish and compare baseline algorithms including both traditional and Reinforecment Learning (RL) approaches. This work sheds insights into the design and development of vision-based TSC approaches and open up new research opportunities. All the code and baselines will be made publicly available.

Saket S. Chaturvedi, Lan Zhang, Wenbin Zhang et al.

3D object detection plays an important role in autonomous driving; however, its vulnerability to backdoor attacks has become evident. By injecting ''triggers'' to poison the training dataset, backdoor attacks manipulate the detector's prediction for inputs containing these triggers. Existing backdoor attacks against 3D object detection primarily poison 3D LiDAR signals, where large-sized 3D triggers are injected to ensure their visibility within the sparse 3D space, rendering them easy to detect and impractical in real-world scenarios. In this paper, we delve into the robustness of 3D object detection, exploring a new backdoor attack surface through 2D cameras. Given the prevalent adoption of camera and LiDAR signal fusion for high-fidelity 3D perception, we investigate the latent potential of camera signals to disrupt the process. Although the dense nature of camera signals enables the use of nearly imperceptible small-sized triggers to mislead 2D object detection, realizing 2D-oriented backdoor attacks against 3D object detection is non-trivial. The primary challenge emerges from the fusion process that transforms camera signals into a 3D space, compromising the association with the 2D trigger to the target output. To tackle this issue, we propose an innovative 2D-oriented backdoor attack against LiDAR-camera fusion methods for 3D object detection, named BadFusion, for preserving trigger effectiveness throughout the entire fusion process. The evaluation demonstrates the effectiveness of BadFusion, achieving a significantly higher attack success rate compared to existing 2D-oriented attacks.

Pan He, Yuxi Chen, Yan Wang et al.

Recently, prompt tuning \cite{lester2021power} has gradually become a new paradigm for NLP, which only depends on the representation of the words by freezing the parameters of pre-trained language models (PLMs) to obtain remarkable performance on downstream tasks. It maintains the consistency of Masked Language Model (MLM) \cite{devlin2018bert} task in the process of pre-training, and avoids some issues that may happened during fine-tuning. Naturally, we consider that the "[MASK]" tokens carry more useful information than other tokens because the model combines with context to predict the masked tokens. Among the current prompt tuning methods, there will be a serious problem of random composition of the answer tokens in prediction when they predict multiple words so that they have to map tokens to labels with the help verbalizer. In response to the above issue, we propose a new \textbf{Pro}mpt \textbf{Tu}ning based on "[\textbf{M}ASK]" (\textbf{Protum}) method in this paper, which constructs a classification task through the information carried by the hidden layer of "[MASK]" tokens and then predicts the labels directly rather than the answer tokens. At the same time, we explore how different hidden layers under "[MASK]" impact on our classification model on many different data sets. Finally, we find that our \textbf{Protum} can achieve much better performance than fine-tuning after continuous pre-training with less time consumption. Our model facilitates the practical application of large models in NLP.

Pan He, Patrick Emami, Sanjay Ranka et al.

Understanding 3D scenes is a critical prerequisite for autonomous agents. Recently, LiDAR and other sensors have made large amounts of data available in the form of temporal sequences of point cloud frames. In this work, we propose a novel problem -- sequential scene flow estimation (SSFE) -- that aims to predict 3D scene flow for all pairs of point clouds in a given sequence. This is unlike the previously studied problem of scene flow estimation which focuses on two frames. We introduce the SPCM-Net architecture, which solves this problem by computing multi-scale spatiotemporal correlations between neighboring point clouds and then aggregating the correlation across time with an order-invariant recurrent unit. Our experimental evaluation confirms that recurrent processing of point cloud sequences results in significantly better SSFE compared to using only two frames. Additionally, we demonstrate that this approach can be effectively modified for sequential point cloud forecasting (SPF), a related problem that demands forecasting future point cloud frames. Our experimental results are evaluated using a new benchmark for both SSFE and SPF consisting of synthetic and real datasets. Previously, datasets for scene flow estimation have been limited to two frames. We provide non-trivial extensions to these datasets for multi-frame estimation and prediction. Due to the difficulty of obtaining ground truth motion for real-world datasets, we use self-supervised training and evaluation metrics. We believe that this benchmark will be pivotal to future research in this area. All code for benchmark and models will be made accessible.

Patrick Emami, Pan He, Sanjay Ranka et al.

Unsupervised multi-object representation learning depends on inductive biases to guide the discovery of object-centric representations that generalize. However, we observe that methods for learning these representations are either impractical due to long training times and large memory consumption or forego key inductive biases. In this work, we introduce EfficientMORL, an efficient framework for the unsupervised learning of object-centric representations. We show that optimization challenges caused by requiring both symmetry and disentanglement can in fact be addressed by high-cost iterative amortized inference by designing the framework to minimize its dependence on it. We take a two-stage approach to inference: first, a hierarchical variational autoencoder extracts symmetric and disentangled representations through bottom-up inference, and second, a lightweight network refines the representations with top-down feedback. The number of refinement steps taken during training is reduced following a curriculum, so that at test time with zero steps the model achieves 99.1% of the refined decomposition performance. We demonstrate strong object decomposition and disentanglement on the standard multi-object benchmark while achieving nearly an order of magnitude faster training and test time inference over the previous state-of-the-art model.

Keke Zhai, Pan He, Tania Banerjee et al.

We propose SparsePipe, an efficient and asynchronous parallelism approach for handling 3D point clouds with multi-GPU training. SparsePipe is built to support 3D sparse data such as point clouds. It achieves this by adopting generalized convolutions with sparse tensor representation to build expressive high-dimensional convolutional neural networks. Compared to dense solutions, the new models can efficiently process irregular point clouds without densely sliding over the entire space, significantly reducing the memory requirements and allowing higher resolutions of the underlying 3D volumes for better performance. SparsePipe exploits intra-batch parallelism that partitions input data into multiple processors and further improves the training throughput with inter-batch pipelining to overlap communication and computing. Besides, it suitably partitions the model when the GPUs are heterogeneous such that the computing is load-balanced with reduced communication overhead. Using experimental results on an eight-GPU platform, we show that SparsePipe can parallelize effectively and obtain better performance on current point cloud benchmarks for both training and inference, compared to its dense solutions.

Xiaohui Huang, Pan He, Anand Rangarajan et al.

In Intelligent Transportation System, real-time systems that monitor and analyze road users become increasingly critical as we march toward the smart city era. Vision-based frameworks for Object Detection, Multiple Object Tracking, and Traffic Near Accident Detection are important applications of Intelligent Transportation System, particularly in video surveillance and etc. Although deep neural networks have recently achieved great success in many computer vision tasks, a uniformed framework for all the three tasks is still challenging where the challenges multiply from demand for real-time performance, complex urban setting, highly dynamic traffic event, and many traffic movements. In this paper, we propose a two-stream Convolutional Network architecture that performs real-time detection, tracking, and near accident detection of road users in traffic video data. The two-stream model consists of a spatial stream network for Object Detection and a temporal stream network to leverage motion features for Multiple Object Tracking. We detect near accidents by incorporating appearance features and motion features from two-stream networks. Using aerial videos, we propose a Traffic Near Accident Dataset (TNAD) covering various types of traffic interactions that is suitable for vision-based traffic analysis tasks. Our experiments demonstrate the advantage of our framework with an overall competitive qualitative and quantitative performance at high frame rates on the TNAD dataset.

Xiaoyong Yuan, Pan He, Xiaolin Andy Li et al.

Recent studies have shown that state-of-the-art deep learning models are vulnerable to the inputs with small perturbations (adversarial examples). We observe two critical obstacles in adversarial examples: (i) Strong adversarial attacks (e.g., C&W attack) require manually tuning hyper-parameters and take a long time to construct an adversarial example, making it impractical to attack real-time systems; (ii) Most of the studies focus on non-sequential tasks, such as image classification, yet only a few consider sequential tasks. In this work, we speed up adversarial attacks, especially on sequential learning tasks. By leveraging the uncertainty of each task, we directly learn the adaptive multi-task weightings, without manually searching hyper-parameters. A unified architecture is developed and evaluated for both non-sequential tasks and sequential ones. To validate the effectiveness, we take the scene text recognition task as a case study. To our best knowledge, our proposed method is the first attempt to adversarial attack for scene text recognition. Adaptive Attack achieves over 99.9\% success rate with 3-6X speedup compared to state-of-the-art adversarial attacks.

Xiaoyu Yue, Zhanghui Kuang, Zhaoyang Zhang et al.

Deep CNNs have achieved great success in text detection. Most of existing methods attempt to improve accuracy with sophisticated network design, while paying less attention on speed. In this paper, we propose a general framework for text detection called Guided CNN to achieve the two goals simultaneously. The proposed model consists of one guidance subnetwork, where a guidance mask is learned from the input image itself, and one primary text detector, where every convolution and non-linear operation are conducted only in the guidance mask. On the one hand, the guidance subnetwork filters out non-text regions coarsely, greatly reduces the computation complexity. On the other hand, the primary text detector focuses on distinguishing between text and hard non-text regions and regressing text bounding boxes, achieves a better detection accuracy. A training strategy, called background-aware block-wise random synthesis, is proposed to further boost up the performance. We demonstrate that the proposed Guided CNN is not only effective but also efficient with two state-of-the-art methods, CTPN and EAST, as backbones. On the challenging benchmark ICDAR 2013, it speeds up CTPN by 2.9 times on average, while improving the F-measure by 1.5%. On ICDAR 2015, it speeds up EAST by 2.0 times while improving the F-measure by 1.0%.

Xiaoyong Yuan, Pan He, Qile Zhu et al.

With rapid progress and significant successes in a wide spectrum of applications, deep learning is being applied in many safety-critical environments. However, deep neural networks have been recently found vulnerable to well-designed input samples, called adversarial examples. Adversarial examples are imperceptible to human but can easily fool deep neural networks in the testing/deploying stage. The vulnerability to adversarial examples becomes one of the major risks for applying deep neural networks in safety-critical environments. Therefore, attacks and defenses on adversarial examples draw great attention. In this paper, we review recent findings on adversarial examples for deep neural networks, summarize the methods for generating adversarial examples, and propose a taxonomy of these methods. Under the taxonomy, applications for adversarial examples are investigated. We further elaborate on countermeasures for adversarial examples and explore the challenges and the potential solutions.

Ruimin Sun, Xiaoyong Yuan, Pan He et al.

Existing malware detectors on safety-critical devices have difficulties in runtime detection due to the performance overhead. In this paper, we introduce PROPEDEUTICA, a framework for efficient and effective real-time malware detection, leveraging the best of conventional machine learning (ML) and deep learning (DL) techniques. In PROPEDEUTICA, all software start execution are considered as benign and monitored by a conventional ML classifier for fast detection. If the software receives a borderline classification from the ML detector (e.g. the software is 50% likely to be benign and 50% likely to be malicious), the software will be transferred to a more accurate, yet performance demanding DL detector. To address spatial-temporal dynamics and software execution heterogeneity, we introduce a novel DL architecture (DEEPMALWARE) for PROPEDEUTICA with multi-stream inputs. We evaluated PROPEDEUTICA with 9,115 malware samples and 1,338 benign software from various categories for the Windows OS. With a borderline interval of [30%-70%], PROPEDEUTICA achieves an accuracy of 94.34% and a false-positive rate of 8.75%, with 41.45% of the samples moved for DEEPMALWARE analysis. Even using only CPU, PROPEDEUTICA can detect malware within less than 0.1 seconds.

Pan He, Weilin Huang, Tong He et al.

We present a novel single-shot text detector that directly outputs word-level bounding boxes in a natural image. We propose an attention mechanism which roughly identifies text regions via an automatically learned attentional map. This substantially suppresses background interference in the convolutional features, which is the key to producing accurate inference of words, particularly at extremely small sizes. This results in a single model that essentially works in a coarse-to-fine manner. It departs from recent FCN- based text detectors which cascade multiple FCN models to achieve an accurate prediction. Furthermore, we develop a hierarchical inception module which efficiently aggregates multi-scale inception features. This enhances local details, and also encodes strong context information, allow- ing the detector to work reliably on multi-scale and multi- orientation text with single-scale images. Our text detector achieves an F-measure of 77% on the ICDAR 2015 bench- mark, advancing the state-of-the-art results in [18, 28]. Demo is available at: http://sstd.whuang.org/.

Zhi Tian, Weilin Huang, Tong He et al.

We propose a novel Connectionist Text Proposal Network (CTPN) that accurately localizes text lines in natural image. The CTPN detects a text line in a sequence of fine-scale text proposals directly in convolutional feature maps. We develop a vertical anchor mechanism that jointly predicts location and text/non-text score of each fixed-width proposal, considerably improving localization accuracy. The sequential proposals are naturally connected by a recurrent neural network, which is seamlessly incorporated into the convolutional network, resulting in an end-to-end trainable model. This allows the CTPN to explore rich context information of image, making it powerful to detect extremely ambiguous text. The CTPN works reliably on multi-scale and multi- language text without further post-processing, departing from previous bottom-up methods requiring multi-step post-processing. It achieves 0.88 and 0.61 F-measure on the ICDAR 2013 and 2015 benchmarks, surpass- ing recent results [8, 35] by a large margin. The CTPN is computationally efficient with 0:14s/image, by using the very deep VGG16 model [27]. Online demo is available at: http://textdet.com/.

Pan He, Weilin Huang, Yu Qiao et al.

We develop a Deep-Text Recurrent Network (DTRN) that regards scene text reading as a sequence labelling problem. We leverage recent advances of deep convolutional neural networks to generate an ordered high-level sequence from a whole word image, avoiding the difficult character segmentation problem. Then a deep recurrent model, building on long short-term memory (LSTM), is developed to robustly recognize the generated CNN sequences, departing from most existing approaches recognising each character independently. Our model has a number of appealing properties in comparison to existing scene text recognition methods: (i) It can recognise highly ambiguous words by leveraging meaningful context information, allowing it to work reliably without either pre- or post-processing; (ii) the deep CNN feature is robust to various image distortions; (iii) it retains the explicit order information in word image, which is essential to discriminate word strings; (iv) the model does not depend on pre-defined dictionary, and it can process unknown words and arbitrary strings. Codes for the DTRN will be available.