论文检索

Yunqian Wen, Bo Liu, Jingyi Cao, Rong Xie, Li Song

Face de-identification involves concealing the true identity of a face while retaining other facial characteristics. Current target-generic methods typically disentangle identity features in the latent space, using adversarial training to balance privacy and utility. However, this pattern often leads to a trade-off between privacy and utility, and the latent space remains difficult to explain. To address these issues, we propose IDeudemon, which employs a "divide and conquer" strategy to protect identity and preserve utility step by step while maintaining good explainability. In Step I, we obfuscate the 3D disentangled ID code calculated by a parametric NeRF model to protect identity. In Step II, we incorporate visual similarity assistance and train a GAN with adjusted losses to preserve image utility. Thanks to the powerful 3D prior and delicate generative designs, our approach could protect the identity naturally, produce high quality details and is robust to different poses and expressions. Extensive experiments demonstrate that the proposed IDeudemon outperforms previous state-of-the-art methods.

Longrong Yang, Xianpan Zhou, Xuewei Li, Liang Qiao, Zheyang Li, Ziwei Yang, Gaoang Wang, Xi Li

Knowledge distillation (KD) has shown potential for learning compact models in dense object detection. However, the commonly used softmax-based distillation ignores the absolute classification scores for individual categories. Thus, the optimum of the distillation loss does not necessarily lead to the optimal student classification scores for dense object detectors. This cross-task protocol inconsistency is critical, especially for dense object detectors, since the foreground categories are extremely imbalanced. To address the issue of protocol differences between distillation and classification, we propose a novel distillation method with cross-task consistent protocols, tailored for the dense object detection. For classification distillation, we address the cross-task protocol inconsistency problem by formulating the classification logit maps in both teacher and student models as multiple binary-classification maps and applying a binary-classification distillation loss to each map. For localization distillation, we design an IoU-based Localization Distillation Loss that is free from specific network structures and can be compared with existing localization distillation losses. Our proposed method is simple but effective, and experimental results demonstrate its superiority over existing methods.

Weiming Zhuang, Yonggang Wen, Lingjuan Lyu, Shuai Zhang

Federated learning (FL) is an emerging distributed machine learning method that empowers in-situ model training on decentralized edge devices. However, multiple simultaneous FL tasks could overload resource-constrained devices. In this work, we propose the first FL system to effectively coordinate and train multiple simultaneous FL tasks. We first formalize the problem of training simultaneous FL tasks. Then, we present our new approach, MAS (Merge and Split), to optimize the performance of training multiple simultaneous FL tasks. MAS starts by merging FL tasks into an all-in-one FL task with a multi-task architecture. After training for a few rounds, MAS splits the all-in-one FL task into two or more FL tasks by using the affinities among tasks measured during the all-in-one training. It then continues training each split of FL tasks based on model parameters from the all-in-one training. Extensive experiments demonstrate that MAS outperforms other methods while reducing training time by 2x and reducing energy consumption by 40%. We hope this work will inspire the community to further study and optimize training simultaneous FL tasks.

Joseph Fioresi, Ishan Rajendrakumar Dave, Mubarak Shah

Video anomaly detection (VAD) without human monitoring is a complex computer vision task that can have a positive impact on society if implemented successfully. While recent advances have made significant progress in solving this task, most existing approaches overlook a critical real-world concern: privacy. With the increasing popularity of artificial intelligence technologies, it becomes crucial to implement proper AI ethics into their development. Privacy leakage in VAD allows models to pick up and amplify unnecessary biases related to people's personal information, which may lead to undesirable decision making. In this paper, we propose TeD-SPAD, a privacy-aware video anomaly detection framework that destroys visual private information in a self-supervised manner. In particular, we propose the use of a temporally-distinct triplet loss to promote temporally discriminative features, which complements current weakly-supervised VAD methods. Using TeD-SPAD, we achieve a positive trade-off between privacy protection and utility anomaly detection performance on three popular weakly supervised VAD datasets: UCF-Crime, XD-Violence, and ShanghaiTech. Our proposed anonymization model reduces private attribute prediction by 32.25% while only reducing frame-level ROC AUC on the UCF-Crime anomaly detection dataset by 3.69%.

Chen Li, Edward G Jones, Steve Furber

This paper presents a new methodology to alleviate the fundamental trade-off between accuracy and latency in spiking neural networks (SNNs). The approach involves decoding confidence information over time from the SNN outputs and using it to develop a decision-making agent that can dynamically determine when to terminate each inference. The proposed method, Dynamic Confidence, provides several significant benefits to SNNs. 1. It can effectively optimize latency dynamically at runtime, setting it apart from many existing low-latency SNN algorithms. Our experiments on CIFAR-10 and ImageNet datasets have demonstrated an average 40% speedup across eight different settings after applying Dynamic Confidence. 2. The decision-making agent in Dynamic Confidence is straightforward to construct and highly robust in parameter space, making it extremely easy to implement. 3. The proposed method enables visualizing the potential of any given SNN, which sets a target for current SNNs to approach. For instance, if an SNN can terminate at the most appropriate time point for each input sample, a ResNet-50 SNN can achieve an accuracy as high as 82.47% on ImageNet within just 4.71 time steps on average. Unlocking the potential of SNNs needs a highly-reliable decision-making agent to be constructed and fed with a high-quality estimation of ground truth. In this regard, Dynamic Confidence represents a meaningful step toward realizing the potential of SNNs.

Sijia Jiang, Jing Hua, Zhizhong Han

In recent years, huge progress has been made on learn- ing neural implicit representations from multi-view images for 3D reconstruction. As an additional input complement- ing coordinates, using sinusoidal functions as positional encodings plays a key role in revealing high frequency de- tails with coordinate-based neural networks. However, high frequency positional encodings make the optimization un- stable, which results in noisy reconstructions and artifacts in empty space. To resolve this issue in a general sense, we introduce to learn neural implicit representations with quantized coordinates, which reduces the uncertainty and ambiguity in the field during optimization. Instead of con- tinuous coordinates, we discretize continuous coordinates into discrete coordinates using nearest interpolation among quantized coordinates which are obtained by discretizing the field in an extremely high resolution. We use discrete coordinates and their positional encodings to learn implicit functions through volume rendering. This significantly re- duces the variations in the sample space, and triggers more multi-view consistency constraints on intersections of rays from different views, which enables to infer implicit function in a more effective way. Our quantized coordinates do not bring any computational burden, and can seamlessly work upon the latest methods. Our evaluations under the widely used benchmarks show our superiority over the state-of-the- art. Our code is available at https://github.com/ MachinePerceptionLab/CQ-NIR.

Zhihao Gu, Liang Liu, Xu Chen, Ran Yi, Jiangning Zhang, Yabiao Wang, Chengjie Wang, Annan Shu, Guannan Jiang, Lizhuang Ma

Knowledge distillation (KD) has been widely explored in unsupervised anomaly detection (AD). The student is assumed to constantly produce representations of typical patterns within trained data, named "normality", and the representation discrepancy between the teacher and student model is identified as anomalies. However, it suffers from the "normality forgetting" issue. Trained on anomaly-free data, the student still well reconstructs anomalous representations for anomalies and is sensitive to fine patterns in normal data, which also appear in training. To mitigate this issue, we introduce a novel Memory-guided Knowledge-Distillation (MemKD) framework that adaptively modulates the normality of student features in detecting anomalies. Specifically, we first propose a normality recall memory (NR Memory) to strengthen the normality of student-generated features by recalling the stored normal information. In this sense, representations will not present anomalies and fine patterns will be well described. Subsequently, we employ a normality embedding learning strategy to promote information learning for the NR Memory. It constructs a normal exemplar set so that the NR Memory can memorize prior knowledge in anomaly-free data and later recall them from the query feature. Consequently, comprehensive experiments demonstrate that the proposed MemKD achieves promising results on five benchmarks, i.e., MVTec AD, VisA, MPDD, MVTec 3D-AD, and Eyecandies.

Taeryung Lee, Yeonguk Oh, Kyoung Mu Lee

In this paper, we propose P3D, the human part-wise motion context learning framework for sign language recognition. Our main contributions lie in two dimensions: learning the part-wise motion context and employing the pose ensemble to utilize 2D and 3D pose jointly. First, our empirical observation implies that part-wise context encoding benefits the performance of sign language recognition. While previous methods of sign language recognition learned motion context from the sequence of the entire pose, we argue that such methods cannot exploit part-specific motion context. In order to utilize part-wise motion context, we propose the alternating combination of a part-wise encoding Transformer (PET) and a whole-body encoding Transformer (WET). PET encodes the motion contexts from a part sequence, while WET merges them into a unified context. By learning part-wise motion context, our P3D achieves superior performance on WLASL compared to previous state-of-the-art methods. Second, our framework is the first to ensemble 2D and 3D poses for sign language recognition. Since the 3D pose holds rich motion context and depth information to distinguish the words, our P3D outperformed the previous state-of-the-art methods employing a pose ensemble.

Guanghui Li, Mingqi Gao, Heng Liu, Xiantong Zhen, Feng Zheng

Referring video object segmentation (RVOS), as a supervised learning task, relies on sufficient annotated data for a given scene. However, in more realistic scenarios, only minimal annotations are available for a new scene, which poses significant challenges to existing RVOS methods. With this in mind, we propose a simple yet effective model with a newly designed cross-modal affinity (CMA) module based on a Transformer architecture. The CMA module builds multimodal affinity with a few samples, thus quickly learning new semantic information, and enabling the model to adapt to different scenarios. Since the proposed method targets limited samples for new scenes, we generalize the problem as - few-shot referring video object segmentation (FS-RVOS). To foster research in this direction, we build up a new FS-RVOS benchmark based on currently available datasets. The benchmark covers a wide range and includes multiple situations, which can maximally simulate real-world scenarios. Extensive experiments show that our model adapts well to different scenarios with only a few samples, reaching state-of-the-art performance on the benchmark. On Mini-Ref-YouTube-VOS, our model achieves an average performance of 53.1 J and 54.8 F, which are 10% better than the baselines. Furthermore, we show impressive results of 77.7 J and 74.8 F on Mini-Ref-SAIL-VOS, which are significantly better than the baselines. Code is publicly available at https://github.com/hengliusky/Few_shot_RVOS.

Zhentao Yu, Zixin Yin, Deyu Zhou, Duomin Wang, Finn Wong, Baoyuan Wang

We introduce a novel framework for one-shot audio-driven talking head generation. Unlike prior works that require additional driving sources for controlled synthesis in a deterministic manner, we instead sample all holistic lip-irrelevant facial motions (i.e. pose, expression, blink, gaze, etc.) to semantically match the input audio while still maintaining both the photo-realism of audio-lip synchronization and overall naturalness. This is achieved by our newly proposed audio-to-visual diffusion prior trained on top of the mapping between audio and non-lip representations. Thanks to the probabilistic nature of the diffusion prior, one big advantage of our framework is it can synthesize diverse facial motion sequences given the same audio clip, which is quite user-friendly for many real applications. Through comprehensive evaluations of public benchmarks, we conclude that (1) our diffusion prior outperforms auto-regressive prior significantly on all the concerned metrics; (2) our overall system is competitive with prior works in terms of audio-lip synchronization but can effectively sample rich and natural-looking lip-irrelevant facial motions while still semantically harmonized with the audio input.

Na Dong, Yongqiang Zhang, Mingli Ding, Gim Hee Lee

Real-world data tends to follow a long-tailed distribution, where the class imbalance results in dominance of the head classes during training. In this paper, we propose a frustratingly simple but effective step-wise learning framework to gradually enhance the capability of the model in detecting all categories of long-tailed datasets. Specifically, we build smooth-tail data where the long-tailed distribution of categories decays smoothly to correct the bias towards head classes. We pre-train a model on the whole long-tailed data to preserve discriminability between all categories. We then fine-tune the class-agnostic modules of the pre-trained model on the head class dominant replay data to get a head class expert model with improved decision boundaries from all categories. Finally, we train a unified model on the tail class dominant replay data while transferring knowledge from the head class expert model to ensure accurate detection of all categories. Extensive experiments on long-tailed datasets LVIS v0.5 and LVIS v1.0 demonstrate the superior performance of our method, where we can improve the AP with ResNet-50 backbone from 27.0% to 30.3% AP, and especially for the rare categories from 15.5% to 24.9% AP. Our best model using ResNet-101 backbone can achieve 30.7% AP, which suppresses all existing detectors using the same backbone.

Guangyi Chen, Xiao Liu, Guangrun Wang, Kun Zhang, Philip H.S. Torr, Xiao-Ping Zhang, Yansong Tang

Video-language pre-trained models have shown remarkable success in guiding video question-answering (VideoQA) tasks. However, due to the length of video sequences, training large-scale video-based models incurs considerably higher costs than training image-based ones. This motivates us to leverage the knowledge from image-based pretraining, despite the obvious gaps between image and video domains. To bridge these gaps, in this paper, we propose Tem-Adapter, which enables the learning of temporal dynamics and complex semantics by a visual Temporal Aligner and a textual Semantic Aligner. Unlike conventional pretrained knowledge adaptation methods that only concentrate on the downstream task objective, the Temporal Aligner introduces an extra language-guided autoregressive task aimed at facilitating the learning of temporal dependencies, with the objective of predicting future states based on historical clues and language guidance that describes event progression. Besides, to reduce the semantic gap and adapt the textual representation for better event description, we introduce a Semantic Aligner that first designs a template to fuse question and answer pairs as event descriptions and then learns a Transformer decoder with the whole video sequence as guidance for refinement. We evaluate Tem-Adapter and different pre-train transferring methods on two VideoQA benchmarks, and the significant performance improvement demonstrates the effectiveness of our method.

Wenqi Ouyang, Yi Dong, Xiaoyang Kang, Peiran Ren, Xin Xu, Xuansong Xie

Retouching images is an essential aspect of enhancing the visual appeal of photos. Although users often share common aesthetic preferences, their retouching methods may vary based on their individual preferences. Therefore, there is a need for white-box approaches that produce satisfying results and enable users to conveniently edit their images simultaneously. Recent white-box retouching methods rely on cascaded global filters that provide image-level filter arguments but cannot perform fine-grained retouching. In contrast, colorists typically employ a divide-and-conquer approach, performing a series of region-specific fine-grained enhancements when using traditional tools like Davinci Resolve. We draw on this insight to develop a white-box framework for photo retouching using parallel region-specific filters, called RSFNet. Our model generates filter arguments (e.g., saturation, contrast, hue) and attention maps of regions for each filter simultaneously. Instead of cascading filters, RSFNet employs linear summations of filters, allowing for a more diverse range of filter classes that can be trained more easily. Our experiments demonstrate that RSFNet achieves state-of-the-art results, offering satisfying aesthetic appeal and increased user convenience for editable white-box retouching.

Ziye Chen, Yu Liu, Mingming Gong, Bo Du, Guoqi Qian, Kate Smith-Miles

State-of-the-art lane detection methods often rely on specific knowledge about lanes -- such as straight lines and parametric curves -- to detect lane lines. While the specific knowledge can ease the modeling process, it poses challenges in handling lane lines with complex topologies (e.g., dense, forked, curved, etc.). Recently, dynamic convolution-based methods have shown promising performance by utilizing the features from some key locations of a lane line, such as the starting point, as convolutional kernels, and convoluting them with the whole feature map to detect lane lines. While such methods reduce the reliance on specific knowledge, the kernels computed from the key locations fail to capture the lane line's global structure due to its long and thin structure, leading to inaccurate detection of lane lines with complex topologies. In addition, the kernels resulting from the key locations are sensitive to occlusion and lane intersections. To overcome these limitations, we propose a transformer-based dynamic kernel generation architecture for lane detection. It utilizes a transformer to generate dynamic convolutional kernels for each lane line in the input image, and then detect these lane lines with dynamic convolution. Compared to the kernels generated from the key locations of a lane line, the kernels generated with the transformer can capture the lane line's global structure from the whole feature map, enabling them to effectively handle occlusions and lane lines with complex topologies. We evaluate our method on three lane detection benchmarks, and the results demonstrate its state-of-the-art performance. Specifically, our method achieves an F1 score of 63.40 on OpenLane and 88.47 on CurveLanes, surpassing the state of the art by 4.30 and 2.37 points, respectively.

Tianyi Shi, Xiaohuan Ding, Liang Zhang, Xin Yang

Curvilinear object segmentation is critical for many applications. However, manually annotating curvilinear objects is very time-consuming and error-prone, yielding insufficiently available annotated datasets for existing supervised methods and domain adaptation methods. This paper proposes a self-supervised curvilinear object segmentation method (FreeCOS) that learns robust and distinctive features from fractals and unlabeled images. The key contributions include a novel Fractal-FDA synthesis (FFS) module and a geometric information alignment (GIA) approach. FFS generates curvilinear structures based on the parametric Fractal L-system and integrates the generated structures into unlabeled images to obtain synthetic training images via Fourier Domain Adaptation. GIA reduces the intensity differences between the synthetic and unlabeled images by comparing the intensity order of a given pixel to the values of its nearby neighbors. Such image alignment can explicitly remove the dependency on absolute intensity values and enhance the inherent geometric characteristics which are common in both synthetic and real images. In addition, GIA aligns features of synthetic and real images via the prediction space adaptation loss (PSAL) and the curvilinear mask contrastive loss (CMCL). Extensive experimental results on four public datasets, i.e., XCAD, DRIVE, STARE and CrackTree demonstrate that our method outperforms the state-of-the-art unsupervised methods, self-supervised methods and traditional methods by a large margin. The source code of this work is available at https://github.com/TY-Shi/FreeCOS.

Nanne van Noord

Dataset summarisation is a fruitful approach to dataset inspection. However, when applied to a single dataset the discovery of visual concepts is restricted to those most prominent. We argue that a comparative approach can expand upon this paradigm to enable richer forms of dataset inspection that go beyond the most prominent concepts. To enable dataset comparison we present a module that learns concept-level prototypes across datasets. We leverage self-supervised learning to discover these prototypes without supervision, and we demonstrate the benefits of our approach in two case-studies. Our findings show that dataset comparison extends dataset inspection and we hope to encourage more works in this direction. Code and usage instructions available at https://github.com/Nanne/ProtoSim

Weiying Zheng, Cheng Xu, Xuemiao Xu, Wenxi Liu, Shengfeng He

Video inpainting aims at filling in missing regions of a video. However, when dealing with dynamic scenes with camera or object movements, annotating the inpainting target becomes laborious and impractical. In this paper, we resolve the one-shot video inpainting problem in which only one annotated first frame is provided. A naive solution is to propagate the initial target to the other frames with techniques like object tracking. In this context, the main obstacles are the unreliable propagation and the partially inpainted artifacts due to the inaccurate mask. For the former problem, we propose curricular inactivation to replace the hard masking mechanism for indicating the inpainting target, which is robust to erroneous predictions in long-term video inpainting. For the latter, we explore the properties of inpainting residue and present an online residue removal method in an iterative detect-and-refine manner. Extensive experiments on several real-world datasets demonstrate the quantitative and qualitative superiorities of our proposed method in one-shot video inpainting. More importantly, our method is extremely flexible that can be integrated with arbitrary traditional inpainting models, activating them to perform the reliable one-shot video inpainting task. Video demonstrations can be found in our supplement, and our code can be found at https://github.com/Arise-zwy/CIRI.

Hang Li, Jindong Gu, Rajat Koner, Sahand Sharifzadeh, Volker Tresp

The field of multimodal research focusing on the comprehension and creation of both images and text has witnessed significant strides. This progress is exemplified by the emergence of sophisticated models dedicated to image captioning at scale, such as the notable Flamingo model and text-to-image generative models, with DALL-E serving as a prominent example. An interesting question worth exploring in this domain is whether Flamingo and DALL-E understand each other. To study this question, we propose a reconstruction task where Flamingo generates a description for a given image and DALL-E uses this description as input to synthesize a new image. We argue that these models understand each other if the generated image is similar to the given image. Specifically, we study the relationship between the quality of the image reconstruction and that of the text generation. We find that an optimal description of an image is one that gives rise to a generated image similar to the original one. The finding motivates us to propose a unified framework to finetune the text-to-image and image-to-text models. Concretely, the reconstruction part forms a regularization loss to guide the tuning of the models. Extensive experiments on multiple datasets with different image captioning and image generation models validate our findings and demonstrate the effectiveness of our proposed unified framework. As DALL-E and Flamingo are not publicly available, we use Stable Diffusion and BLIP in the remaining work. Project website: https://dalleflamingo.github.io.

Gabriele Berton, Gabriele Trivigno, Barbara Caputo, Carlo Masone

Visual Place Recognition is a task that aims to predict the place of an image (called query) based solely on its visual features. This is typically done through image retrieval, where the query is matched to the most similar images from a large database of geotagged photos, using learned global descriptors. A major challenge in this task is recognizing places seen from different viewpoints. To overcome this limitation, we propose a new method, called EigenPlaces, to train our neural network on images from different point of views, which embeds viewpoint robustness into the learned global descriptors. The underlying idea is to cluster the training data so as to explicitly present the model with different views of the same points of interest. The selection of this points of interest is done without the need for extra supervision. We then present experiments on the most comprehensive set of datasets in literature, finding that EigenPlaces is able to outperform previous state of the art on the majority of datasets, while requiring 60% less GPU memory for training and using 50% smaller descriptors. The code and trained models for EigenPlaces are available at https://github.com/gmberton/EigenPlaces, while results with any other baseline can be computed with the codebase at https://github.com/gmberton/auto_VPR.

Xuanyu Yi, Jiajun Deng, Qianru Sun, Xian-Sheng Hua, Joo-Hwee Lim, Hanwang Zhang

We tackle the data scarcity challenge in few-shot point cloud recognition of 3D objects by using a joint prediction from a conventional 3D model and a well-pretrained 2D model. Surprisingly, such an ensemble, though seems trivial, has hardly been shown effective in recent 2D-3D models. We find out the crux is the less effective training for the "joint hard samples", which have high confidence prediction on different wrong labels, implying that the 2D and 3D models do not collaborate well. To this end, our proposed invariant training strategy, called INVJOINT, does not only emphasize the training more on the hard samples, but also seeks the invariance between the conflicting 2D and 3D ambiguous predictions. INVJOINT can learn more collaborative 2D and 3D representations for better ensemble. Extensive experiments on 3D shape classification with widely-adopted ModelNet10/40, ScanObjectNN and Toys4K, and shape retrieval with ShapeNet-Core validate the superiority of our INVJOINT.