Peng Wu

Peng Wu

Auto-generated profile, based on publicly available information

Peng Wu

Northwestern Polytechnical University

Research Focus & Metrics

Academic Career Length

11Years (2016-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
24 March 2024 in Proceedings of the AAAI Conference on Artificial Intelligence
Verified
Journal Article
Open access
PDF file
PWPeng Wu
Peng Wu
XZXuerong Zhou
Xuerong Zhou
GPGuansong Pang
Guansong Pang
LZLingru Zhou
Lingru Zhou
QYQingsen Yan
Qingsen Yan
YZYanning Zhang
Yanning Zhang
The recent contrastive language-image pre-training (CLIP) model has shown great success in a wide range of image-level tasks, revealing remarkable ability for learning powerful visual representations with rich semantics. An open and worthwhile problem is efficiently adapting such a strong model to the video domain and designing a robust video anomaly detector. In this work, we propose VadCLIP, a new paradigm for weakly supervised video anomaly detection (WSVAD) by leveraging the frozen CLIP model directly without any pre-training and fine-tuning process. Unlike current works that directly feed extracted features into the weakly supervised classifier for frame-level binary classification, VadCLIP makes full use of fine-grained associations between vision and language on the strength of CLIP and involves dual branch. One branch simply utilizes visual features for coarse-grained binary classification, while the other fully leverages the fine-grained language-image alignment. With the benefit of dual branch, VadCLIP achieves both coarse-grained and fine-grained video anomaly detection by transferring pre-trained knowledge from CLIP to WSVAD task. We conduct extensive experiments on two commonly-used benchmarks, demonstrating that VadCLIP achieves the best performance on both coarse-grained and fine-grained WSVAD, surpassing the state-of-the-art methods by a large margin. Specifically, VadCLIP achieves 84.51% AP and 88.02% AUC on XD-Violence and UCF-Crime, respectively. Code and features are released at https://github.com/nwpu-zxr/VadCLIP.

3 March 2021 in IEEE Transactions on Image Processing
Verified
Journal Article
Closed access
PWPeng Wu
Peng Wu
JLJing Liu
Jing Liu
Weakly supervised anomaly detection is a challenging task since frame-level labels are not given in the training phase. Previous studies generally employ neural networks to learn features and produce frame-level predictions and then use multiple instance learning (MIL)-based classification loss to ensure the interclass separability of the learned features; all operations simply take into account the current time information as input and ignore the historical observations. According to investigations, these solutions are universal but ignore two essential factors, i.e., the temporal cue and feature discrimination. The former introduces temporal context to enhance the current time feature, and the latter enforces the samples of different categories to be more separable in the feature space. In this article, we propose a method that consists of four modules to leverage the effect of these two ignored factors. The causal temporal relation (CTR) module captures local-range temporal dependencies among features to enhance features. The classifier (CL) projects enhanced features to the category space using the causal convolution and further expands the temporal modeling range. Two additional modules, namely, compactness (CP) and dispersion (DP) modules, are designed to learn the discriminative power of features, where the compactness module ensures the intraclass compactness of normal features, and the dispersion module enhances the interclass dispersion. Extensive experiments on three public benchmarks demonstrate the significance of causal temporal relations and feature discrimination for anomaly detection and the superiority of our proposed method.

5 September 2019 in IEEE Transactions on Neural Networks and Learning Systems
Verified
Journal Article
Closed access
PWPeng Wu
Peng Wu
JLJing Liu
Jing Liu
FSFang Shen
Fang Shen
How to build a generic deep one-class (DeepOC) model to solve one-class classification problems for anomaly detection, such as anomalous event detection in complex scenes? The characteristics of existing one-class labels lead to a dilemma: it is hard to directly use a multiple classifier based on deep neural networks to solve one-class classification problems. Therefore, in this article, we propose a novel DeepOC neural network, termed as DeepOC, which can simultaneously learn compact feature representations and train a DeepOC classifier. Only with the given normal samples, we use the stacked convolutional encoder to generate their low-dimensional high-level features and train a one-class classifier to make these features as compact as possible. Meanwhile, for the sake of the correct mapping relation and the feature representations' diversity, we utilize a decoder in order to reconstruct raw samples from these low-dimensional feature representations. This structure is gradually established using an adversarial mechanism during the training stage. This mechanism is the key to our model. It organically combines two seemingly contradictory components and allows them to take advantage of each other, thus making the model robust and effective. Unlike methods that use handcrafted features or those that are separated into two stages (extracting features and training classifiers), DeepOC is a one-stage model using reliable features that are automatically extracted by neural networks. Experiments on various benchmark data sets show that DeepOC is feasible and achieves the state-of-the-art anomaly detection results compared with a dozen existing methods.

1 February 2022 in IEEE Transactions on Multimedia
Verified
Journal Article
Closed access
PWPeng Wu
Peng Wu
XLXiaotao Liu
Xiaotao Liu
JLJing Liu
Jing Liu
Violence detection in video is very promising in practical applications due to the emergence of massive videos in recent years. Most previous works define violence detection as a simple video classification task and use the single modality of small-scale datasets, e.g., visual signal. However, such solutions are undersupplied. To mitigate this problem, we first introduce a large-scale audio-visual XD-Violence dataset, which is composed of 4754 untrimmed videos with a total duration of 217 hours. Then we introduce two complementary tasks, i.e., coarse-grained violent frame detection and fine-grained violent event detection, to advance the simple violence video classification to frame-level violent event localization, which aims to accurately locate the violent events on untrimmed videos. We propose a novel network that takes as input audio-visual information and contains three parallel branches to capture different relationships among video snippets and further integrate features, where similarity branch and proximity branch capture long-range dependencies using similarity prior and proximity prior respectively, and score branch dynamically captures the closeness of predicted score. Our method outperforms other state-of-the-art methods on two benchmarks. Moreover, experiment results also show the positive effect of audio-visual input and relationship modeling. The dataset and code are released in https://roc-ng.github.io/XD-Violence/.

1 November 2020 in Pattern Recognition
Verified
Journal Article
Closed access
PWPeng Wu
Peng Wu
JLJing Liu
Jing Liu
MLMingming Li
Mingming Li
YSYujia Sun
Yujia Sun
FSFang Shen
Fang Shen

18 March 2024 in IEEE Transactions on Image Processing
Verified
Journal Article
Closed access
PWPeng Wu
Peng Wu
JLJing Liu
Jing Liu
XHXiangteng He
Xiangteng He
YPYuxin Peng
Yuxin Peng
PWPeng Wang
Peng Wang
YZYanning Zhang
Yanning Zhang
Video anomaly detection (VAD) has been paid increasing attention due to its potential applications, its current dominant tasks focus on online detecting anomalies, which can be roughly interpreted as the binary or multiple event classification. However, such a setup that builds relationships between complicated anomalous events and single labels, e.g., “vandalism”, is superficial, since single labels are deficient to characterize anomalous events. In reality, users tend to search a specific video rather than a series of approximate videos. Therefore, retrieving anomalous events using detailed descriptions is practical and positive but few researches focus on this. In this context, we propose a novel task called Video Anomaly Retrieval (VAR), which aims to pragmatically retrieve relevant anomalous videos by cross-modalities, e.g., language descriptions and synchronous audios. Unlike the current video retrieval where videos are assumed to be temporally well-trimmed with short duration, VAR is devised to retrieve long untrimmed videos which may be partially relevant to the given query. To achieve this, we present two large-scale VAR benchmarks and design a model called Anomaly-Led Alignment Network (ALAN) for VAR. In ALAN, we propose an anomaly-led sampling to focus on key segments in long untrimmed videos. Then, we introduce an efficient pretext task to enhance semantic associations between video-text fine-grained representations. Besides, we leverage two complementary alignments to further match cross-modal contents. Experimental results on two benchmarks reveal the challenges of VAR task and also demonstrate the advantages of our tailored method. Captions are publicly released at https://github.com/Roc-Ng/VAR .