暗DSAR:用于暗视频中的动作识别的轻量级单步管道
Yuwei Yin1, Miao Liu2, Renjie Yang3
1The State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan, China.
概括
一种新的基于变压器的方法,暗域转换动作识别 (Dark-DSAR),提高了暗视频动作识别的准确性. 这种单步方法减少了计算,同时提高了对具有挑战性的数据集的性能.
科学领域:
- 计算机视觉 计算机视觉
- 人工智能的人工智能
- 机器学习 机器学习
背景情况:
- 在黑暗视频中识别人为动作对于现实应用至关重要.
- 由于忽视了环境背景,现有的方法在黑暗场景中难以准确.
- 目前的两步方法很复杂,而一步方法缺乏效率.
研究的目的:
- 开发一个高效和准确的单步方法,用于暗视频人类动作识别.
- 解决现有处理黑暗环境的方法的局限性.
- 为了提高域迁移和分类任务之间的功能连贯性.
主要方法:
- 提出了一种基于变压器的单步方法,名为"暗域转移以进行动作识别" (Dark-DSAR).
- 集成的域迁移和分类在单个步骤中使用域转移模块 (DSM) 来进行暗到明的适应.
- 通过探索视频大小和模型匹配,包括空间分辨率下降,优化推断效率.
主要成果:
- 在ARID1.5 (89.49%) 上实现了最先进的Top-1精度,超过现有方法2.56%.
- 在HMDB51-Dark (67.13%) 和UAV-human-night (61.9%) 上表现强.
- 废除研究表明,DSM至少提高了1%的动作分类器准确度.
结论:
- 暗视频DSAR为暗视频动作识别提供了一个计算效率高但高度准确的解决方案.
- 拟议的DSM有效地将模型适应黑暗条件,提高识别性能.
- 该方法在具有挑战性的低光场景中为动作识别提供了重大进展.
相关概念视频
02:16Light Acquisition
8.0K
In order to produce glucose, plants need to capture sufficient light energy. Many modern plants have evolved leaves specialized for light acquisition. Leaves can be only millimeters in width or tens of meters wide, depending on the environment. Due to competition for sunlight, evolution has driven the evolution of increasingly larger leaves and taller plants, to avoid shading by their neighbors with contaminant elaboration of root architecture and mechanisms to transport water and nutrients.
8.0K
01:10Detection of Black Holes
1.7K
Although black holes were theoretically postulated in the 1920s, they remained outside the domain of observational astronomy until the 1970s.
Their closest cousins are neutron stars, which are composed almost entirely of neutrons packed against each other, making them extremely dense. A neutron star has the same mass as the Sun but its diameter is only a few kilometers. Therefore, the escape velocity from their surface is close to the speed of light.
Not until the 1960s, when the first neutron...
Their closest cousins are neutron stars, which are composed almost entirely of neutrons packed against each other, making them extremely dense. A neutron star has the same mass as the Sun but its diameter is only a few kilometers. Therefore, the escape velocity from their surface is close to the speed of light.
Not until the 1960s, when the first neutron...
1.7K
01:22Force Classification
2.8K
Forces play a crucial role in the study of physics and engineering. They are essential in describing the motion, behavior, and equilibrium of objects in the physical world. Forces can be classified based on their origin, type, and direction of action.
Contact and non-contact forces are two of the most widely used categories of forces. As the name suggests, contact forces require physical contact between two objects to act upon each other. Examples of contact forces include frictional,...
Contact and non-contact forces are two of the most widely used categories of forces. As the name suggests, contact forces require physical contact between two objects to act upon each other. Examples of contact forces include frictional,...
2.8K
01:05Difference from Background: Limit of Detection
9.2K
The limit of detection (LOD) is the smallest amount of analyte that can be distinguished from the background noise. The LOD value corresponds to the concentration at which the analyte signal is three times larger than the standard deviation of the blank signal. Below this value, the analyte signal cannot be differentiated from the background noise. It is calculated by dividing the calibration slope by 3 times the standard deviation of the blank signals.
The LOD indicates the presence or absence...
The LOD indicates the presence or absence...
9.2K
01:20Deconvolution
776
Deconvolution, also known as inverse filtering, is the process of extracting the impulse response from known input and output signals. This technique is vital in scenarios where the system's characteristics are unknown, and they must be inferred from the observable signals.
Deconvolution involves several mathematical techniques to derive the impulse response. One common approach is polynomial division. In this method, the input and output sequences are treated as coefficients of...
Deconvolution involves several mathematical techniques to derive the impulse response. One common approach is polynomial division. In this method, the input and output sequences are treated as coefficients of...
776