PERSON DETECTION, TRACKING, RE-IDENTIFICATION AND SEARCH:
|
Person detection, tracking and re-identification in videos is an important problem in computer vision
thanks to its wide applications in various video analysis scenarios.
As a result, it has attracted huge interest from the scientific community.
Our work in this topic aims at developing a fully automatic person detection, tracking and re-identification.
Read more about this work
|
PERSON SEARCH THROUGH VIETNAMESE NATURAL LANGUAGE:
|
Nowadays, surveillance camera systems are widely deployed today from public places to private houses.
This leads to huge image databases. Recent years have witnessed a significant improvement of surveillance video analysis,
especially for person detection and tracking. However, finding the interested person in these databases is still very challenging issue.
A majority of the existing person search methods bases on the assumption that the example image of the person of interest is available.
This assumption is however not always satisfied in practical situations.
In this work, we focus on person search method with query of Vietnamese natural language.
Read more about this work
|
SURVEILLANCE VIDEO INDEXING AND RETRIEVAL:
|
In this topic, we are interested in modeling mobile objects in surveillance video and in matching these objects.
In surveillance video, a mobile object such as person can be detected and tracked in several frames. In each frame, this object is represented by a blob (minimum bounding box). In order to retrieve this object, it is necessary to extract the representative blobs and to define the matching function of two mobile objects.
For this, we have proposed different representative blob extraction methods and a matching method based on EMD (Earth Mover Distance)
Read more about this topic
|
METHEMATICAL EXPRESSION DETECTION AND RECOGNITION IN SCIENTIFIC DOCUMENT IMAGES:
|
Mathematical expressions have been widely used in scientific documents.
In order to analyze the documents, automatic detection of mathematical expressions is a crucial step.
The main aim of our work is to develop methods for understanding mathematical expressions in scientific document images.
To this end, two main tasks have been conducted: mathematical expression (inline and isolated expression) detection and mathematical expression recognition.
Read more about this work
|
PLANT IDENTIFICATION AND RETRIEVAL:
|
The main aim of this topic is to develop new methods for plant species identification from images.
In this topic, we have developed:
- A new leaf-based plant identification from complex background images
- Fusion schemes for multi-organ based plant identification
- Plant organ detection/classification
- An application for Vietnamese medicinal plant retrieval
Read more about this work
|
HUMAN POSTURE, HUMAN ACTIVITY AND ABNORMAL EVENT RECOGNITION:
|
In this work, we are interested on modeling and recognizing human posture, human activity and abnormal event from multimodal information (e.g., color, depth, skeleton, accelerometer).
Read more about this work
|
HAND POSTURE AND GESTURE RECOGNITION:
|
Nowadays, people want to interact with machines more naturally. One of the powerful communication channels is hand gesture.
Vision-based approach has involved many researchers because this approach does not require any extra device.
One of the key problems we need to resolve is hand posture recognition on RGB images because it can be used directly or integrated into a multi-cues hand gesture recognition.
The main challenges of this problem are illumination differences, cluttered background, background changes, high intra-class variation, and high inter-class similarity.
In this topic, we focus on hand detection and hand posture recognition.
In hand detection step, we employed Viola-Jones detector with proposed concept Internal Haarlike feature.
The proposed hand detection works in real-time within frames captured from real complex environments and avoids unexpected effects of background.
The proposed detector outperforms original Viola-Jones detector using traditional Haar-like feature.
In hand posture recognition step, we proposed a new hand representation based on a good generic descriptor that is kernel descriptor (KDES).
Read more about this work
|
3D OBJECT DETECTION, LOCALIZATION AND RECOGNITION:
|
The goal is to detect, recognize and describe objects from range data.
The detection is restricted to round objects such as cups, glasses, bottles, pencils, and rectangular objects, e.g., a small box.
The recognition has to be made fast and reliable, and it must give a sense of direct feeling
Read more about this work
|