Liquid Content Detection In Transparent Containers: A Benchmark

Abstract

Various substances that possess liquid states include drinking water, various types of fuel, pharmaceuticals, and chemicals, which are indispensable in our daily lives. There are numerous real-world applications for liquid content detection in transparent containers, for example, service robots, pouring robots, security checks, industrial observation systems, etc. However, the majority of the existing methods either concentrate on transparent container detection or liquid height estimation; the former provides very limited information for more advanced computer vision tasks, whereas the latter is too demanding to generalize to open-world applications. In this paper, we propose a dataset for detecting liquid content in transparent containers (LCDTC), which presents an innovative task involving transparent container detection and liquid content estimation. The primary objective of this task is to obtain more information beyond the location of the container by additionally providing certain liquid content information which is easy to achieve with computer vision methods in various open-world applications. This task has potential applications in service robots, waste classification, security checks, and so on. The presented LCDTC dataset comprises 5916 images that have been extensively annotated through axis-aligned bounding boxes. We develop two baseline detectors, termed LCD-YOLOF and LCD-YOLOX, for the proposed dataset, based on two identity-preserved human posture detectors, i.e., IPH-YOLOF and IPH-YOLOX. By releasing LCDTC, we intend to stimulate more future works into the detection of liquid content in transparent containers and bring more focus to this challenging task.

PatchRLNet: A Framework Combining a Vision Transformer and Reinforcement Learning for The Separation of a PTFE Emulsion and Paraffin

“…In [25], Hongliang Li et al demonstrated a liquid recognition scheme based on the vision transformer network, which combines an optical flow-controlled refractive index sensor with visual intelligence algorithms and does not require a spectrometer and precise metasurface mediation, further demonstrating the role of the vision transformer network in liquid-state detection. In addition, You Wu et al [26] proposed a dataset for detecting liquid content in transparent containers (LCDTC), leading to an innovative task involving transparent container detection and liquid content estimation. The dataset proposed by the author developed two baseline detectors, called LCD-YOLOF and LCD-YOLOX, and further proposed a Swin Transformer Integration (WISTE) method for automatically identifying the water index of water bodies.…”

Beyond Human Detection: A Benchmark for Detecting Common Human Posture

“…Additionally, the constant λ is used as the weight coefficient to represent the loss for the prediction head of common human postures. We follow [47,83,84] to provide the definitions of these losses below:…”

Design, simulation, control of a hybrid pouring robot: enhancing automation level in the foundry industry

“…However, task-based optimization is needed to determine the appropriate parameters for the best performance. In the research based on the combination of depth vision and liquid level, Do Chau [27], Cheng DF [28], and Wu Y [29] carried out related research, but the casting performance and flexibility need to be improved. Audio information can partially compensate for the lack of visual information and enhance the generalization ability of robot pouring skill.…”