While neural fields have made significant strides in view synthesis and scene reconstruction, editing them poses a formidable challenge due to their implicit encoding of geometry and texture information from multi-vie...
详细信息
Self-supervised skeleton-based action recognition has attracted more attention in recent years. By utilizing the unlabeled data, more generalizable features can be learned to alleviate the overfitting problem and redu...
Self-supervised skeleton-based action recognition has attracted more attention in recent years. By utilizing the unlabeled data, more generalizable features can be learned to alleviate the overfitting problem and reduce the demand for massive labeled training data. Inspired by the MAE [1], we propose a spatial-temporal masked autoencoder framework for self-supervised 3D skeleton-based action recognition (SkeletonMAE). Following MAE's masking and reconstruction pipeline, we utilize a skeleton-based encoder-decoder transformer architecture to reconstruct the masked skeleton sequences. A novel masking strategy, named Spatial-Temporal Masking, is introduced in terms of both joint-level and frame-level for the skeleton sequence. This pre-training strategy makes the encoder output generalizable skeleton features with spatial and temporal dependencies. Given the unmasked skeleton sequence, the encoder is fine-tuned for the action recognition task. Extensive experiments show that our SkeletonMAE achieves remarkable performance and outperforms the state-of-the-art methods on both NTU RGB+D 60 and NTU RGB+D 120 datasets.
The usage of online entertainment has increased dramatically after some time with the enhancement of the Internet and has turned into the most compelling systems administration stage in this century. Notwithstanding, ...
详细信息
Event data can asynchronously capture variations in light intensity, thereby implicitly providing valuable complementary cues for RGB-Event tracking. Existing methods typically employ a direct interaction mechanism to...
详细信息
As the concept of smart cities continues to gain momentum, the integration of various technologies has become essential for creating sustainable and efficient urban environments. However, with the proliferation of int...
详细信息
With the increase in computer use for the last 10 years, it increases the size of cyber-attacks as well. Since most computers are connected to the network, they may suffer cyber-attacks. Because of these cyber threats...
详细信息
Machine learning has the ability to dramatically improve sustainable systems by anticipating needs, maximizing resource use, raising output, and lowering waste. An overview of earlier studies on the incorporation of m...
详细信息
Data-free model stealing aims to replicate a target model without direct access to either the training data or the target model. To accomplish this, existing methods use a generator to produce samples in order to trai...
详细信息
Human-Robot Interaction (HRI) becomes more and more important in a world where robots integrate fast in all aspects of our lives but HRI applications depend massively on the utilized robotic system as well as the depl...
详细信息
Smart cities are cities that are designed to be more efficient, sustainable, and connected. As our cities grow and become more complex, it's important to find new ways to address the challenges that arise, such as...
详细信息
暂无评论