ICCV 2023 论文列表

Existing methods of privacy-preserving action recognition (PPAR) mainly focus on frame-level (spatial) privacy removal through 2D CNNs. Unfortunately, they have two major drawbacks. First, they may compromise temporal dynamics in input videos, which are critical for accurate action recognition. Second, they are vulnerable to practical attacking scenarios where attackers probe for privacy from an entire video rather than individual frames. To address these issues, we propose a novel framework STPrivacy to perform video-level PPAR. For the first time, we introduce vision Transformers into PPAR by treating a video as a tubelet sequence, and accordingly design two complementary mechanisms, i.e., sparsification and anonymization, to remove privacy from a spatio-temporal perspective. In specific, our privacy sparsification mechanism applies adaptive token selection to abandon action-irrelevant tubelets. Then, our anonymization mechanism implicitly manipulates the remaining action-tubelets to erase privacy in the embedding space through adversarial learning. These mechanisms provide significant advantages in terms of privacy preservation for human eyes and action-privacy trade-off adjustment during deployment. We additionally contribute the first two large-scale PPAR benchmarks, VP-HMDB51 and VP-UCF101, to the community. Extensive evaluations on them, as well as two other tasks, validate the effectiveness and generalization capability of our framework.

We address the problem of extracting key steps from unlabeled procedural videos, motivated by the potential of Augmented Reality (AR) headsets to revolutionize job training and performance. We decompose the problem into two steps: representation learning and key steps extraction. We propose a training objective, Bootstrapped Multi-Cue Contrastive (BMC2) loss to learn discriminative representations for various steps without any labels. Different from prior works, we develop techniques to train a light-weight temporal module which uses off-the-shelf features for self supervision. Our approach can seamlessly leverage information from multiple cues like optical flow, depth or gaze to learn discriminative features for key-steps, making it amenable for AR applications. We finally extract key steps via a tunable algorithm that clusters the representations and samples. We show significant improvements over prior works for the task of key step localization and phase classification. Qualitative results demonstrate that the extracted key steps are meaningful and succinctly represent various steps of the procedural tasks.

Scale variation is a deep-rooted problem in object counting, which has not been effectively addressed by existing scale-aware algorithms. An important factor is that they typically involve cooperative learning across multi-resolutions, which could be suboptimal for learning the most discriminative features from each scale. In this paper, we propose a novel method termed STEERER (SelecTivE inhERitance lEaRning) that addresses the issue of scale variations in object counting. STEERER selects the most suitable scale for patch objects to boost feature extraction and only inherits discriminative features from lower to higher resolution progressively. The main insights of STEERER are a dedicated Feature Selection and Inheritance Adaptor (FSIA), which selectively forwards scale-customized features at each scale, and a Masked Selection and Inheritance Loss (MSIL) that helps to achieve high-quality density maps across all scales. Our experimental results on nine datasets with counting and localization tasks demonstrate the unprecedented scale generalization ability of STEERER. Code is available at https://github.com/taohan10200/STEERER.

Surrogate gradient (SG) is one of the most effective approaches for training spiking neural networks (SNNs). While assisting SNNs to achieve classification performance comparable to artificial neural networks, SG suffers from the problem of time-consuming training, preventing it from efficient learning. In this paper, we formally analyze the backward process of classic SG and find that the membrane accumulation through time leads to exponential growth of training time. With this discovery, we propose Stabilized Spiking Flow (SSF), a simple yet effective approach to accelerate training of SG-based SNNs. For each spiking neuron, SSF averages its input and output activations over time to yield stabilized input and output, respectively. Then, instead of back propagating all errors that are related to current neuron and inherently entangled in time domain, the auxiliary gradient is directly propagated from the stabilized output to input through a devised relationship mapping. Additionally, SSF method is suitable to different neuron models. Extensive experiments on both static and neuromorphic datasets demonstrate that SNNs trained with SSF approach can achieve performance comparable to the original counterparts, while reducing the training time significantly. In particular, SSF speeds up the training process of state-of-the-art SNN models up to 10x when time steps equal to 80.

Source-free domain adaptation (SFDA) is a popular unsupervised domain adaptation method where a pre-trained model from a source domain is adapted to a target domain without accessing any source data. Despite rich results in this area, existing literature overlooks the security challenges of the unsupervised SFDA setting in presence of a malicious source domain owner. This work investigates the effect of a source adversary which may inject a hidden malicious behavior (Backdoor/Trojan) during source training and potentially transfer it to the target domain even after benign training by the victim (target domain owner). Our investigation of the current SFDA setting reveals that because of the unique challenges present in SFDA (e.g., no source data, target label), defending against backdoor attack using existing defenses become practically ineffective in protecting the target model. To address this, we propose a novel target domain protection scheme called secure source-free domain adaptation (SSDA). SSDA adopts a single-shot model compression of a pre-trained source model and a novel knowledge transfer scheme with a spectral-norm-based loss penalty for target training. The proposed static compression and the dynamic training loss penalty are designed to suppress the malicious channels responsive to the backdoor during the adaptation stage. At the same time, the knowledge transfer from an uncompressed auxiliary model helps to recover the benign test accuracy. Our extensive evaluation on multiple dataset and domain tasks against recent backdoor attacks reveal that the proposed SSDA can successfully defend against strong backdoor attacks with little to no degradation in test accuracy compared to the vulnerable baseline SFDA methods. Our code is available at https://github.com/ML-Security-Research-LAB/SSDA.

Semi-supervised learning (SSL) methods effectively leverage unlabeled data to improve model generalization. However, SSL models often underperform in open-set scenarios, where unlabeled data contain outliers from novel categories that do not appear in the labeled set. In this paper, we study the challenging and realistic open-set SSL setting, where the goal is to both correctly classify inliers and to detect outliers. Intuitively, the inlier classifier should be trained on inlier data only. However, we find that inlier classification performance can be largely improved by incorporating high-confidence pseudo-labeled data, regardless of whether they are inliers or outliers. Also, we propose to utilize non-linear transformations to separate the features used for inlier classification and outlier detection in the multi-task learning framework, preventing adverse effects between them. Additionally, we introduce pseudo-negative mining, which further boosts outlier detection performance. The three ingredients lead to what we call Simple but Strong Baseline (SSB) for open-set SSL. In experiments, SSB greatly improves both inlier classification and outlier detection performance, outperforming existing methods by a large margin. Our code will be released at https://github.com/YUE-FAN/SSB.

Previous works have shown that increasing the window size for Transformer-based image super-resolution models (e.g., SwinIR) can significantly improve the model performance but the computation overhead is also considerable. In this paper, we present SRFormer, a simple but novel method that can enjoy the benefit of large window self-attention but introduces even less computational burden. The core of our SRFormer is the permuted self-attention(PSA), which strikes an appropriate balance between the channel and spatial information for self-attention. Our PSA is simple and can be easily applied to existing super-resolution networks based on window self-attention. Without any bells and whistles, we show that our SRFormer achieves a 33.86dB PSNR score on the Urban100 dataset, which is 0.46dB higher than that of SwinIR but uses fewer parameters and computations. We hope our simple and effective approach can serve as a useful tool for future research in super-resolution model design. Our code is available at https://github.com/HVision-NKU/SRFormer.

Smartphone photography is becoming increasingly popular, but fitting high-performing camera systems within the given space limitations remains a challenge for manufacturers. As a result, powerful mobile camera systems are in high demand. Despite recent progress in computer vision, camera system quality assessment remains a tedious and manual process. In this paper, we present the Smartphone Camera Quality Assessment Dataset (SQAD), which includes natural images captured by 29 devices. SQAD defines camera system quality based on six widely accepted criteria: resolution, color accuracy, noise level, dynamic range, Point Spread Function, and aliasing. Built on thorough examinations in a controlled laboratory environment, SQAD provides objective metrics for quality assessment, overcoming previous subjective opinion scores. Moreover, we introduce the task of automatic camera quality assessment and train deep learning-based models on the collected data to perform a precise quality prediction for arbitrary photos. The dataset, codes and pre-trained models are released at https://github.com/aiff22/SQAD.

Recent studies show that self-attentions behave like low-pass filters (as opposed to convolutions) and enhancing their high-pass filtering capability improves model performance. Contrary to this idea, we investigate existing convolution-based models with spectral analysis and observe that improving the low-pass filtering in convolution operations also leads to performance improvement. To account for this observation, we hypothesize that utilizing optimal token mixers that capture balanced representations of both high- and low-frequency components can enhance the performance of models. We verify this by decomposing visual features into the frequency domain and combining them in a balanced manner. To handle this, we replace the balancing problem with a mask filtering problem in the frequency domain. Then, we introduce a novel token-mixer named SPAM and leverage it to derive a MetaFormer model termed as SPANet. Experimental results show that the proposed method provides a way to achieve this balance, and the balanced representations of both high- and low-frequency components can improve the performance of models on multiple computer vision tasks. Our code is available at https://doranlyong.github.io/projects/spanet/.

Animating portraits using speech has received growing attention in recent years, with various creative and practical use cases. An ideal generated video should have good lip sync with the audio, natural facial expressions and head motions, and high frame quality. In this work, we present SPACE, which uses speech and a single image to generate high-resolution, and expressive videos with realistic head pose, without requiring a driving video. It uses a multi-stage approach, combining the controllability of facial landmarks with the high-quality synthesis power of a pretrained face generator. SPACE also allows for the control of emotions and their intensities. Our method outperforms prior methods in objective metrics for image quality and facial motions and is strongly preferred by users in pair-wise comparisons. Please visit the project page to view the videos and to see more results: https://research.nvidia.com/labs/dir/space.

Most learning-based approaches to category-level 6D pose estimation are design around normalized object coordinate space (NOCS). While being successful, NOCS-based methods become inaccurate and less robust when handling objects of a category containing significant intra-category shape variations. This is because the object coordinates induced by global and rigid alignment of objects are semantically incoherent, making the coordinate regression hard to learn and generalize. We propose Semantically-aware Object Coordinate Space (SOCS) built by warping-and-aligning the objects guided by a sparse set of keypoints with semantically meaningful correspondence. SOCS is semantically coherent: Any point on the surface of a object can be mapped to a semantically meaningful location in SOCS, allowing for accurate pose and size estimation under large shape variations. To learn effective coordinate regression to SOCS, we propose a novel multi-scale coordinate-based attention network. Evaluations demonstrate that our method is easy to train, well-generalizing for large intra-category shape variations and robust to inter-object occlusions.

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

CutMix is a vital augmentation strategy that determines the performance and generalization ability of vision transformers (ViTs). However, the inconsistency between the mixed images and the corresponding labels harms its efficacy. Existing CutMix variants tackle this problem by generating more consistent mixed images or more precise mixed labels, but inevitably introduce heavy training overhead or require extra information, undermining ease of use. To this end, we propose an efficient and effective Self-Motivated image Mixing method (SMMix), which motivates both image and label enhancement by the model under training itself. Specifically, we propose a max-min attention region mixing approach that enriches the attention-focused objects in the mixed images. Then, we introduce a fine-grained label assignment technique that co-trains the output tokens of mixed images with fine-grained supervision. Moreover, we devise a novel feature consistency constraint to align features from mixed and unmixed images. Due to the subtle designs of the self-motivated paradigm, our SMMix is significant in its smaller training overhead and better performance than other CutMix variants. In particular, SMMix improves the accuracy of DeiT-T/S/B, CaiT-XXS-24/36, and PVT-T/S/M/L by more than +1% on ImageNet-1k. The generalization capability of our method is also demonstrated on downstream tasks and out-of-distribution datasets. Our project is available at https://github.com/ChenMnZ/SMMix.

Video-language pre-training is crucial for learning powerful multi-modal representation. However, it typically requires a massive amount of computation. In this paper, we develop SMAUG, an efficient pre-training framework for video-language models. The foundation component in SMAUG is masked autoencoders. Different from prior works which only mask textual inputs, our masking strategy considers both visual and textual modalities, providing a better cross-modal alignment and saving more pre-training costs. On top of that, we introduce a space-time token sparsification module, which leverages context information to further select only "important" spatial regions and temporal frames for pre-training. Coupling all these designs allows our method to enjoy both competitive performances on text-to-video retrieval and video question answering tasks, and much less pre-training costs by 1.9x or more. For example, our SMAUG only needs 50 NVIDIA A6000 GPU hours for pre-training to attain competitive performances on these two video-language tasks across six popular benchmarks.

The goal of continual learning is to improve the performance of recognition models in learning sequentially arrived data. Although most existing works are established on the premise of learning from scratch, growing efforts have been devoted to incorporating the benefits of pre-training. However, how to adaptively exploit the pre-trained knowledge for each incremental task while maintaining its generalizability remains an open question. In this work, we present an extensive analysis for continual learning on a pre-trained model (CLPM), and attribute the key challenge to a progressive overfitting problem. Observing that selectively reducing the learning rate can almost resolve this issue in the representation layer, we propose a simple but extremely effective approach named Slow Learner with Classifier Alignment (SLCA), which further improves the classification layer by modeling the class-wise distributions and aligning the classification layers in a post-hoc fashion. Across a variety of scenarios, our proposal provides substantial improvements for CLPM (e.g., up to 49.76%, 50.05%, 44.69% and 40.16% on Split CIFAR-100, Split ImageNet-R, Split CUB-200 and Split Cars-196, respectively), and thus outperforms state-of-the-art approaches by a large margin. Based on such a strong baseline, critical factors and promising directions are analyzed in-depth to facilitate subsequent research.

Learning fine-grained interplay between vision and language contributes to a more accurate understanding for Vision-Language tasks. However, it remains challenging to extract key image regions according to the texts for semantic alignments. Most existing works are either limited by text-agnostic and redundant regions obtained with the frozen detectors, or failing to scale further due to their heavy reliance on scarce grounding (gold) data to pre-train detectors. To solve these problems, we propose Self-Locator Aided Network (SLAN) for vision-language understanding tasks without any extra gold data. SLAN consists of a region filter and a region adaptor to localize regions of interest conditioned on different texts. By aggregating vision-language information, the region filter selects key regions and the region adaptor updates their coordinates with text guidance. With detailed region-word alignments, SLAN can be easily generalized to many downstream tasks. It achieves fairly competitive results on five vision-language understanding tasks (e.g., 85.7% and 69.2% on COCO image-to-text and text-to-image retrieval, surpassing previous SOTA methods). SLAN also demonstrates strong zero-shot and fine-tuned transferability to two localization tasks.

This paper introduces SKiT, a fast Key information Transformer for phase recognition of videos. Unlike previous methods that rely on complex models to capture long-term temporal information, SKiT accurately recognizes high-level stages of videos using an efficient key pooling operation. This operation records important key information by retaining the maximum value recorded from the beginning up to the current video frame, with a time complexity of O(1). Experimental results on Cholec80 and AutoLaparo surgical datasets demonstrate the ability of our model to recognize phases in an online manner. SKiT achieves higher performance than state-of-the-art methods with an accuracy of 92.5% and 82.9% on Cholec80 and AutoLaparo, respectively, while running the temporal model eight times faster ( 7ms v.s. 55ms) than LoViT, which uses ProbSparse to capture global information. We highlight that the inference time of SKiT is constant, and independent from the input length, making it a stable choice for keeping a record of important global information, that appears on long surgical videos, essential for phase recognition. To sum up, we propose an effective and efficient model for surgical phase recognition that leverages key global information. This has an intrinsic value when performing this task in an online manner on long surgical videos for stable real-time surgical recognition systems.

Text-to-image diffusion models are gradually introduced into computer graphics, recently enabling the development of Text-to-3D pipelines in an open domain. However, for interactive editing purposes, local manipulations of content through a simplistic textual interface can be arduous. Incorporating user guided sketches with Text-to-image pipelines offers users more intuitive control. Still, as state-of-the-art Text-to-3D pipelines rely on optimizing Neural Radiance Fields (NeRF) through gradients from arbitrary rendering views, conditioning on sketches is not straightforward. In this paper, we present SKED, a technique for editing 3D shapes represented by NeRFs. Our technique utilizes as few as two guiding sketches from different views to alter an existing neural field. The edited region respects the prompt semantics through a pre-trained diffusion model. To ensure the generated output adheres to the provided sketches, we propose novel loss functions to generate the desired edits while preserving the density and radiance of the base instance. We demonstrate the effectiveness of our proposed method through several qualitative and quantitative experiments. https://sked-paper.github.io/

Point cloud registration is essential for many applications. However, existing real datasets require extremely tedious and costly annotations, yet may not provide accurate camera poses. For the synthetic datasets, they are mainly object-level, so the trained models may not generalize well to real scenes. We design SIRA-PCR, a new approach to 3D point cloud registration. First, we build a synthetic scene-level 3D registration dataset, specifically designed with physically-based and random strategies to arrange diverse objects. Second, we account for variations in different sensing mechanisms and layout placements, then formulate a sim-to-real adaptation framework with an adaptive re-sample module to simulate patterns in real point clouds. To our best knowledge, this is the first work that explores sim-to-real adaptation for point cloud registration. Extensive experiments show the SOTA performance of SIRA-PCR on widely-used indoor and outdoor datasets. The code and dataset will be released on https://github.com/Chen-Suyi/SIRA_Pytorch.