Zero-shot recognition of generic objects
| Project/Area Number |
19K24344
|
|---|
| Research Category |
Grant-in-Aid for Research Activity Start-up
|
| Allocation Type | Multi-year Fund |
|---|
| Review Section |
1001:Information science, computer engineering, and related fields
|
|---|
| Research Institution | Kobe University |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2019-08-30 – 2022-03-31
|
| Project Status |
Completed (Fiscal Year 2021)
|
| Budget Amount *help |
¥2,860,000 (Direct Cost: ¥2,200,000、Indirect Cost: ¥660,000)
Fiscal Year 2020: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
Fiscal Year 2019: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
|
|---|
| Keywords | Zero-Shot Learning / Self-Supervised Learning / Visual Representation / Feature Extraction / Semantic representations / Resource Efficiency / CNN / Computer vision / Language Models / Object recognition / Deep learning / Computational efficiency / Semantic representation |
|---|
| Outline of Research at the Start |
This study focuses on deriving new principles for optimization and semantic feature learning applied to generic object recognition. On the optimization front, we will focus on improving the computational and algorithmic efficiency of training deep learning models. In order to expand our search space by enabling quicker iteration over different architectural designs. On the semantic learning front, we aim to achieve a better understanding of the visual features that can be derived from semantic data, which we believe to be the key missing element to enable practical Zero-Shot recognition.
|
| Outline of Final Research Achievements |
Until a few years ago, computers could not recognize things in pictures. For example, computers were not capable to tell whether any human is in a given picture or not. Around ten years ago, computer programs became capable to recognize a number of things in pictures with high precision, including humans, dogs, cars, etc. The development of many technologies such as self-driving vehicles and robots were previously limited by the inability of computers to recognize such objects: for example, a self-driving car can not drive if it can not recognize a pedestrian on the road.
However, computers can currently only recognize a finite number of things such as "a man" or "a woman", while humans can recognize things with more details and nuance such as "a young asian woman on a bike". This research project has worked towards giving computers the ability to recognize more complex and less predefined things, in order to allow computers to take better decisions.
|
| Academic Significance and Societal Importance of the Research Achievements |
本研究では、写真から不特定の物の秋類を認識するために必要な情報を研究しました。 視覚情報の特定の処理は、意味表現よりも重要であり、そのようなプログラムを生成する能力は、実行できる計算の量によって制限されることがわかりました。
|
Report
(4 results)
Research Products
(4 results)
