Humans watch more than a billion hours of video per day. Most of this video was edited manually, which is a tedious process. However, AI-enabled video-generation and video-editing is on the rise. Building on text-to-image models like Stable Diffusion and Imagen, generative AI has improved dramatically on video tasks. But it's hard to evaluate progress in these video tasks because there is no standard benchmark. So, we propose a new dataset for text-guided video editing (TGVE), and we run a competition at CVPR to evaluate models on our TGVE dataset. In this paper we present a retrospective on the competition and describe the winning method. The competition dataset is available at https://sites.google.com/view/loveucvpr23/track4.
Training recommendation models on large datasets often requires significant time and computational resources. Consequently, an emergent imperative has arisen to construct informative, smaller-scale datasets for efficiently training. Dataset compression techniques explored in other domains show potential possibility to address this problem, via sampling a subset or synthesizing a small dataset. However, applying existing approaches to condense recommendation datasets is impractical due to following challenges: (i) sampling-based methods are inadequate in addressing the long-tailed distribution problem; (ii) synthesizing-based methods are not applicable due to discreteness of interactions and large size of recommendation datasets; (iii) neither of them fail to address the specific issue in recommendation of false negative items, where items with potential user interest are incorrectly sampled as negatives owing to insufficient exposure. To bridge this gap, we investigate dataset condensation for recommendation, where discrete interactions are continualized with probabilistic re-parameterization. To avoid catastrophically expensive computations, we adopt a one-step update strategy for inner model training and introducing policy gradient estimation for outer dataset synthesis. To mitigate amplification of long-tailed problem, we compensate long-tailed users in the condensed dataset. Furthermore, we propose to utilize a proxy model to identify false negative items. Theoretical analysis regarding the convergence property is provided. Extensive experiments on multiple datasets demonstrate the efficacy of our method. In particular, we reduce the dataset size by 75% while approximating over 98% of the original performance on Dianping and over 90% on other datasets.
The Generic Event Boundary Detection (GEBD) task aims to build a model for segmenting videos into segments by detecting general event boundaries applicable to various classes. In this paper, based on last year's MAE-GEBD method, we have improved our model performance on the GEBD task by adjusting the data processing strategy and loss function. Based on last year's approach, we extended the application of pseudo-label to a larger dataset and made many experimental attempts. In addition, we applied focal loss to concentrate more on difficult samples and improved our model performance. Finally, we improved the segmentation alignment strategy used last year, and dynamically adjusted the segmentation alignment method according to the boundary density and duration of the video, so that our model can be more flexible and fully applicable in different situations. With our method, we achieve an F1 score of 86.03% on the Kinetics-GEBD test set, which is a 0.09% improvement in the F1 score compared to our 2022 Kinetics-GEBD method.
In multi-domain learning (MDL) scenarios, high labeling effort is required due to the complexity of collecting data from various domains. Active Learning (AL) presents an encouraging solution to this issue by annotating a smaller number of highly informative instances, thereby reducing the labeling effort. Previous research has relied on conventional AL strategies for MDL scenarios, which underutilize the domain-shared information of each instance during the selection procedure. To mitigate this issue, we propose a novel perturbation-based two-stage multi-domain active learning (P2S-MDAL) method incorporated into the well-regarded ASP-MTL model. Specifically, P2S-MDAL involves allocating budgets for domains and establishing regions for diversity selection, which are further used to select the most cross-domain influential samples in each region. A perturbation metric has been introduced to evaluate the robustness of the shared feature extractor of the model, facilitating the identification of potentially cross-domain influential samples. Experiments are conducted on three real-world datasets, encompassing both texts and images. The superior performance over conventional AL strategies shows the effectiveness of the proposed strategy. Additionally, an ablation study has been carried out to demonstrate the validity of each component. Finally, we outline several intriguing potential directions for future MDAL research, thus catalyzing the field's advancement.
Large Language Models (LLMs) have demonstrated exceptional performance in a variety of tasks, including essay writing and question answering. However, it is crucial to address the potential misuse of these models, which can lead to detrimental outcomes such as plagiarism and spamming. Recently, several detectors have been proposed, including fine-tuned classifiers and various statistical methods. In this study, we reveal that with the aid of carefully crafted prompts, LLMs can effectively evade these detection systems. We propose a novel Substitution-based In-Context example Optimization method (SICO) to automatically generate such prompts. On three real-world tasks where LLMs can be misused, SICO successfully enables ChatGPT to evade six existing detectors, causing a significant 0.54 AUC drop on average. Surprisingly, in most cases these detectors perform even worse than random classifiers. These results firmly reveal the vulnerability of existing detectors. Finally, the strong performance of SICO suggests itself as a reliable evaluation protocol for any new detector in this field.
Multi-domain learning (MDL) refers to simultaneously constructing a model or a set of models on datasets collected from different domains. Conventional approaches emphasize domain-shared information extraction and domain-private information preservation, following the shared-private framework (SP models), which offers significant advantages over single-domain learning. However, the limited availability of annotated data in each domain considerably hinders the effectiveness of conventional supervised MDL approaches in real-world applications. In this paper, we introduce a novel method called multi-domain contrastive learning (MDCL) to alleviate the impact of insufficient annotations by capturing both semantic and structural information from both labeled and unlabeled data.Specifically, MDCL comprises two modules: inter-domain semantic alignment and intra-domain contrast. The former aims to align annotated instances of the same semantic category from distinct domains within a shared hidden space, while the latter focuses on learning a cluster structure of unlabeled instances in a private hidden space for each domain. MDCL is readily compatible with many SP models, requiring no additional model parameters and allowing for end-to-end training. Experimental results across five textual and image multi-domain datasets demonstrate that MDCL brings noticeable improvement over various SP models.Furthermore, MDCL can further be employed in multi-domain active learning (MDAL) to achieve a superior initialization, eventually leading to better overall performance.
Autonomous exploration is a crucial aspect of robotics that has numerous applications. Most of the existing methods greedily choose goals that maximize immediate reward. This strategy is computationally efficient but insufficient for overall exploration efficiency. In recent years, some state-of-the-art methods are proposed, which generate a global coverage path and significantly improve overall exploration efficiency. However, global optimization produces high computational overhead, leading to low-frequency planner updates and inconsistent planning motion. In this work, we propose a novel method to support fast UAV exploration in large-scale and cluttered 3-D environments. We introduce a computationally low-cost viewpoints generation method using novel occlusion-free spheres. Additionally, we combine greedy strategy with global optimization, which considers both computational and exploration efficiency. We benchmark our method against state-of-the-art methods to showcase its superiority in terms of exploration efficiency and computational time. We conduct various real-world experiments to demonstrate the excellent performance of our method in large-scale and cluttered environments.
Data heterogeneity is an inherent challenge that hinders the performance of federated learning (FL). Recent studies have identified the biased classifiers of local models as the key bottleneck. Previous attempts have used classifier calibration after FL training, but this approach falls short in improving the poor feature representations caused by training-time classifier biases. Resolving the classifier bias dilemma in FL requires a full understanding of the mechanisms behind the classifier. Recent advances in neural collapse have shown that the classifiers and feature prototypes under perfect training scenarios collapse into an optimal structure called simplex equiangular tight frame (ETF). Building on this neural collapse insight, we propose a solution to the FL's classifier bias problem by utilizing a synthetic and fixed ETF classifier during training. The optimal classifier structure enables all clients to learn unified and optimal feature representations even under extremely heterogeneous data. We devise several effective modules to better adapt the ETF structure in FL, achieving both high generalization and personalization. Extensive experiments demonstrate that our method achieves state-of-the-art performances on CIFAR-10, CIFAR-100, and Tiny-ImageNet.
Tracking multiple athletes in sports videos is a very challenging Multi-Object Tracking (MOT) task, since athletes often have the same appearance and are intimately covered with each other, making a common occlusion problem becomes an abhorrent duplicate detection. In this paper, the duplicate detection is newly and precisely defined as occlusion misreporting on the same athlete by multiple detection boxes in one frame. To address this problem, we meticulously design a novel transformer-based Duplicate Detection Decontaminator (D$^3$) for training, and a specific algorithm Rally-Hungarian (RH) for matching. Once duplicate detection occurs, D$^3$ immediately modifies the procedure by generating enhanced boxes losses. RH, triggered by the team sports substitution rules, is exceedingly suitable for sports videos. Moreover, to complement the tracking dataset that without shot changes, we release a new dataset based on sports video named RallyTrack. Extensive experiments on RallyTrack show that combining D$^3$ and RH can dramatically improve the tracking performance with 9.2 in MOTA and 4.5 in HOTA. Meanwhile, experiments on MOT-series and DanceTrack discover that D$^3$ can accelerate convergence during training, especially save up to 80 percent of the original training time on MOT17. Finally, our model, which is trained only with volleyball videos, can be applied directly to basketball and soccer videos for MAT, which shows priority of our method. Our dataset is available at https://github.com/heruihr/rallytrack.
Generic Event Boundary Detection (GEBD) tasks aim at detecting generic, taxonomy-free event boundaries that segment a whole video into chunks. In this paper, we apply Masked Autoencoders to improve algorithm performance on the GEBD tasks. Our approach mainly adopted the ensemble of Masked Autoencoders fine-tuned on the GEBD task as a self-supervised learner with other base models. Moreover, we also use a semi-supervised pseudo-label method to take full advantage of the abundant unlabeled Kinetics-400 data while training. In addition, we propose a soft-label method to partially balance the positive and negative samples and alleviate the problem of ambiguous labeling in this task. Lastly, a tricky segmentation alignment policy is implemented to refine boundaries predicted by our models to more accurate locations. With our approach, we achieved 85.94% on the F1-score on the Kinetics-GEBD test set, which improved the F1-score by 2.31% compared to the winner of the 2021 Kinetics-GEBD Challenge. Our code is available at https://github.com/ContentAndMaterialPortrait/MAE-GEBD.