Theses and Projects: Difference between revisions

(24 intermediate revisions by 9 users not shown)

Line 21:

* (P) Student project

=== * '''New!''' Tree Growth Detection using Satellite Images and Computer Vision Methods (B/M/P) ===

=== * '''New!''' Task Offloading and Resource Allocation Optimization===

With the continuous advancement of the next-generation wireless communication technologies and the population of mobile devices, a variety of Internet of Things (IoT) applications are emerging and seeking efficient task execution paradigms. This topic presents efficient joint task offloading and auction-based resource allocation mechanisms in edge computing, which not only expand the computational capabilities of mobile devices but also enhance the Quality of Service of IoT applications by significantly reducing latency. We expect you have a background in edge computing, optimization algorithms, and programming skills.

Please contact Dongkuo Wu [dongkuo.wu@cs.uni-goettingen.de]

=== * '''New!''' Efficient Live Volumetric Video Streaming System===

The exponential growth of digital data and multimedia content necessitates robust and efficient systems to handle the streaming of high-resolution, three-dimensional volumetric videos. These videos offer a more immersive and realistic experience, making them increasingly used in various sectors such as virtual reality, augmented reality, and entertainment. The challenge here lies in creating a system that can handle the high-bandwidth and computation-intensive demands of live volumetric video streaming while ensuring the delivery of a seamless and high-quality user experience. This project conceptualizes the development and optimization of efficient algorithms and systems to handle volumetric video streams, mitigating bandwidth cost and latency issues. We expect you to have a background in video streaming technologies, computer vision, and programming skills.

Please contact Yanlong Huang[yanlong.huang@cs.uni-goettingen.de]

=== * '''New!''' Edge-Cloud Orchestration for LiDAR-based Traffic Analysis===

The imminent era of smart cities and autonomous vehicles paves the way for the deployment and operation of advanced monitoring and processing systems. Among these, LiDAR technology stands out for its ability to provide high-resolution, three-dimensional traffic data, becoming an essential component for efficient traffic analysis and management. However, the computation-intensive and latency-sensitive nature of LiDAR data processing poses significant challenges and dictates the need for efficient orchestration between edge and cloud computing resources. Edge-Cloud Orchestration offers an innovative solution to this problem by bridging the gap between these two technologies, enabling the low-latency processing of complex LiDAR data. It would be good if you have a background in point cloud processing/cloud computing, K8s, and programming skills.

Please contact Yanlong Huang[yanlong.huang@cs.uni-goettingen.de]

=== * '''New!''' Using LLM for Sign Language Translation (B/M/P)===

Sign language is the primary means of communication for the deaf and hard-of-hearing community, yet most people do not understand it. This topic explores the integration of Large Language Models (LLMs) with computer vision to build an advanced sign language translation system - with special focus on overcoming the critical challenge of understanding long, continuous sign language videos. We welcome students passionate about Natural Language Processing (NLP) and Computer Vision (CV) to explore the cutting edge of sign language translation technology.

Please contact Wenfang Wu [wenfang.wu@cs.uni-goettingen.de]

=== * '''New!''' Using LLM for Sentiment Knowledge Graph Construction (B/M/P)===

Constructing a sentiment knowledge graph using Large Language Models (LLMs) like ChatGPT involves leveraging the model's capabilities to understand and analyze textual data, extract entities and relationships, perform sentiment analysis, and organize the information into a graph structure. We expect you have a background in knowledge graph and programming skills in Python.

Please contact Wenfang Wu [wenfang.wu@cs.uni-goettingen.de]

=== * '''New!''' Using LLM for Knowledge Graph Completion (B/M/P)===

Large language models (LLMs), such as ChatGPT and GPT-4 (OpenAI, 2023), have extensive internal knowledge repositories from their vast pretraining corpora, which can be used as an extra knowledge base to alleviate information scarcity for the long-tail entities in Knowledge Graphs. However, there is no effective workflow design for LLM on KGC tasks. How to leverage the LLM to perform reasoning on the KG Completion (KGC) task is a noteworthy and significant topic. We need students for this topic. We expect you to have a background in knowledge graph and LLMs, you'd better have a programming skill in Python.

Please contact Tong Shen [shen.tong@cs.uni-goettingen.de]

=== * '''New!''' Context Specific Self-supervised Pre-Training for Remote Sensing Applications (Semantic Segmentation, Change Detection, Socio-Economic Indicator Estimation, ...) (B/M/P) ===

Satellite images in combination with Machine/Deep Learning models have shown to be an effective tool for analysis and monitoring tasks regarding disasters, deforestation, climate change, socio-economic estimation and others. The training of these models usually rely on labelled ground-truth data, which is labour intensive and therefore often scarcely available. To overcome this limitation, models are often trained in self-supervised approaches with unlabelled data, such as Contrastive Learning or Masked Autoencoders. However, these approaches are completely independent and not related to the intended downstream task. In this project/thesis the relationship between the pre-training task and the model performance on the downstream task will be explored and self-supervised training approaches tailored for a selected remote sensing downstream task (semantic segmentation of trees, tree crown share per pixel estimation, change detection of disasters, socio-economic estimation, ...) will be developed.

Please contact Fabian Wölk [fabian.woelk@cs.uni-goettingen.de]

=== [Occupied] Tree Growth Detection using Satellite Images and Computer Vision Methods (B/M/P) ===

A tree planting project in Madagascar was initiated several years ago. The outcomes of this project shall now be evaluated by analyzing satellite images of the study area with Computer Vision methods. In a first step, very high resolution (VHR) satellite images from 2023 with a resolution of 0.5m will be used to identify trees with object detection / semantic segmentation. In the next step a lower resolution (5m) satellite image time series starting in 2015 will be used for change detection to identify, in which locations the project was (un)successful.

Line 27:

Line 71:

Please contact Fabian Wölk [fabian.woelk@cs.uni-goettingen.de]

=== * '''New!''' Image-to-Image Translation of Different Nightlight Image Types (B/M/P) ===

=== * '''New!''' Emotional Support Conversation Generation based on LLM (B/M/P)===

Emotional support conversation aims to reduce individuals' emotional distress through social interaction and help them understand and cope with the challenges they face. Using LLM to provide emotional support is a promising technology which can be used in customer service chats, mental health support and so on. We need students for this topic. We expect you have a background in dialogue generation and programming skills in Python.

Please contact Jing Li [jing.li@cs.uni-goettingen.de]

=== * '''New!''' Rumor control and detection method for social networks based on GCN===

In social networks, rumors spread quickly and have a wide impact. Through GCN, the complex relationships between nodes in the network can be effectively captured, and the detection and propagation paths of rumors can be modeled and controlled. The core of this method is to improve the ability to identify and control the spread of rumors by jointly learning user behavior, information content, and network topology by building an information propagation graph. We need students in this topic. We expect you have a background in rumor detection and programming skills in Python.

Please contact Fei Gao [fei.gao@cs.uni-goettingen.de]

=== [Occupied] Image-to-Image Translation of Different Nightlight Image Types (B/M/P) ===

Nightlight intensities have been proven to be a good indicator for socio-economic status. However, for long-term temporal analyses their use can be challenging, as different satellites for sensing nightlight intensities operated at different times (DMSP OLS 1992-2014 and VIIRS 2012-2023). Both types differ not only in resolution, but there is also a big discrepancy in the optical appearance and value ranges. To obtain consistent nightlight images for temporal analysis, Image-to-Image Translation methods shall be used in this project/thesis for the conversion between both types. Finally the performance of the translated and original nightlight images for a regression on socio-economic indicators shall be evaluated.

Line 69:

Line 125:

Please contact Dr. Tingting Yuan [tingting.yuan@cs.uni-goettingen.de]] (B/M/P)

=== * Privacy-preserved Video Analytics===

=== * '''[Closed]''' Privacy-preserved Video Analytics===

This project/thesis topic focuses on the protection of privacy in video analytics.

Line 84:

Line 140:

[1] Pecam: privacy-enhanced video streaming and analytics via securely-reversible transformation [https://dl.acm.org/doi/abs/10.1145/3447993.3448618].

=== * AI for networking adaption ===

=== * '''[Closed]''' AI for networking adaption ===

In this project/theses topic, you will explore how to make AI meets networking requirements (e.g., fluctuating network states).

You will (1) deploy and test Genet[1]; (2)extend the Genet environment to multi-client environment (e.g., ABR); (3) deploy multi-agent algorithms on Genet and valid the performance.

Line 109:

Line 165:

Please contact Dr. Tingting Yuan [tingting.yuan@cs.uni-goettingen.de] (B/M/P)

=== * '''~~New!~~''' Socioecomonic analysis based on spatiotemporal and linguistic analysis on microblogging data ===

=== * '''[Closed]''' Socioecomonic analysis based on spatiotemporal and linguistic analysis on microblogging data ===

Identifying the socioeconomic status (SES) of users in social media like Twitter or Weibo is useful e.g., for digitized advertisements and social policies. This study aims to collect profiles of Twitter users on selected topics such as culture or foreign language learning, extract the temporal, spatial and linguistic features, and compare different classification algorithms (e.g., decision tree, random forest, na\"{i}ve Bayes, deep learning, and Gaussian processes classifier) to predict the socioeconomic status.

Line 118:

Line 174:

Please contact Prof. Xiaoming Fu [fu@cs.uni-goettingen.de](B/M/P)

=== [Closed] Super resolution technique for efficient video delivery ===

Line 157:

Line 212:

Please contact Dr.Tingting Yuan [tingting.yuan@cs.uni-goettingen.de ] and Weijun Wang [weijun.wang@informatik.uni-goettingen.de](B/M/P)

~~-->~~

== Ongoing Topics ==

Line 186:

Line 238:

| Completed by Jason

|-

| ~~Video~~ analytics ~~with deep reinforcement learning~~ (Master Thesis)

| Neural video analytics(Master Thesis)

|[http://www.net.informatik.uni-goettingen.de/?q=people/dr-tingting-yuan Tingting Yuan]

|

@@ Line 21: / Line 21: @@
 * (P) Student project
-===  * '''New!''' Tree Growth Detection using Satellite Images and Computer Vision Methods (B/M/P) ===
+===  * '''New!''' Task Offloading and Resource Allocation Optimization===
+With the continuous advancement of the next-generation wireless communication technologies and the population of mobile devices, a variety of Internet of Things (IoT) applications are emerging and seeking efficient task execution paradigms. This topic presents efficient joint task offloading and auction-based resource allocation mechanisms in edge computing, which not only expand the computational capabilities of mobile devices but also enhance the Quality of Service of IoT applications by significantly reducing latency. We expect you have a background in edge computing, optimization algorithms, and programming skills.
+Please contact Dongkuo Wu [dongkuo.wu@cs.uni-goettingen.de]
+===  * '''New!''' Efficient Live Volumetric Video Streaming System===
+The exponential growth of digital data and multimedia content necessitates robust and efficient systems to handle the streaming of high-resolution, three-dimensional volumetric videos. These videos offer a more immersive and realistic experience, making them increasingly used in various sectors such as virtual reality, augmented reality, and entertainment. The challenge here lies in creating a system that can handle the high-bandwidth and computation-intensive demands of live volumetric video streaming while ensuring the delivery of a seamless and high-quality user experience. This project conceptualizes the development and optimization of efficient algorithms and systems to handle volumetric video streams, mitigating bandwidth cost and latency issues. We expect you to have a background in video streaming technologies, computer vision, and programming skills.
+Please contact Yanlong Huang[yanlong.huang@cs.uni-goettingen.de]
+===  * '''New!''' Edge-Cloud Orchestration for LiDAR-based Traffic Analysis===
+The imminent era of smart cities and autonomous vehicles paves the way for the deployment and operation of advanced monitoring and processing systems. Among these, LiDAR technology stands out for its ability to provide high-resolution, three-dimensional traffic data, becoming an essential component for efficient traffic analysis and management. However, the computation-intensive and latency-sensitive nature of LiDAR data processing poses significant challenges and dictates the need for efficient orchestration between edge and cloud computing resources. Edge-Cloud Orchestration offers an innovative solution to this problem by bridging the gap between these two technologies, enabling the low-latency processing of complex LiDAR data. It would be good if you have a background in point cloud processing/cloud computing, K8s, and programming skills.
+Please contact Yanlong Huang[yanlong.huang@cs.uni-goettingen.de]
+===  * '''New!''' Using LLM for Sign Language Translation (B/M/P)===
+Sign language is the primary means of communication for the deaf and hard-of-hearing community, yet most people do not understand it. This topic explores the integration of Large Language Models (LLMs) with computer vision to build an advanced sign language translation system - with special focus on overcoming the critical challenge of understanding long, continuous sign language videos. We welcome students passionate about Natural Language Processing (NLP) and Computer Vision (CV) to explore the cutting edge of sign language translation technology.
+Please contact Wenfang Wu [wenfang.wu@cs.uni-goettingen.de]
+===  * '''New!''' Using LLM for Sentiment Knowledge Graph Construction (B/M/P)===
+Constructing a sentiment knowledge graph using Large Language Models (LLMs) like ChatGPT involves leveraging the model's capabilities to understand and analyze textual data, extract entities and relationships, perform sentiment analysis, and organize the information into a graph structure. We expect you have a background in knowledge graph and programming skills in Python.
+Please contact Wenfang Wu [wenfang.wu@cs.uni-goettingen.de]
+===  * '''New!''' Using LLM for Knowledge Graph Completion (B/M/P)===
+Large language models (LLMs), such as ChatGPT and GPT-4 (OpenAI, 2023), have extensive internal knowledge repositories from their vast pretraining corpora, which can be used as an extra knowledge base to alleviate information scarcity for the long-tail entities in Knowledge Graphs. However, there is no effective workflow design for LLM on KGC tasks. How to leverage the LLM to perform reasoning on the KG Completion (KGC) task is a noteworthy and significant topic. We need students for this topic. We expect you to have a background in knowledge graph and LLMs, you'd better have a programming skill in Python.
+Please contact Tong Shen [shen.tong@cs.uni-goettingen.de]
+===  * '''New!''' Context Specific Self-supervised Pre-Training for Remote Sensing Applications (Semantic Segmentation, Change Detection, Socio-Economic Indicator Estimation, ...) (B/M/P) ===
+Satellite images in combination with Machine/Deep Learning models have shown to be an effective tool for analysis and monitoring tasks regarding disasters, deforestation, climate change, socio-economic estimation and others. The training of these models usually rely on labelled ground-truth data, which is labour intensive and therefore often scarcely available. To overcome this limitation, models are often trained in self-supervised approaches with unlabelled data, such as Contrastive Learning or Masked Autoencoders. However, these approaches are completely independent and not related to the intended downstream task. In this project/thesis the relationship between the pre-training task and the model performance on the downstream task will be explored and self-supervised training approaches tailored for a selected remote sensing downstream task (semantic segmentation of trees, tree crown share per pixel estimation, change detection of disasters, socio-economic estimation, ...) will be developed.
+Please contact Fabian Wölk [fabian.woelk@cs.uni-goettingen.de]
+===  [Occupied] Tree Growth Detection using Satellite Images and Computer Vision Methods (B/M/P) ===
 A tree planting project in Madagascar was initiated several years ago. The outcomes of this project shall now be evaluated by analyzing satellite images of the study area with Computer Vision methods. In a first step, very high resolution (VHR) satellite images from 2023 with a resolution of 0.5m will be used to identify trees with object detection / semantic segmentation. In the next step a lower resolution (5m) satellite image time series starting in 2015 will be used for change detection to identify, in which locations the project was  (un)successful.
@@ Line 27: / Line 71: @@
 Please contact Fabian Wölk [fabian.woelk@cs.uni-goettingen.de]
-===  * '''New!''' Image-to-Image Translation of Different Nightlight Image Types (B/M/P) ===
+===  * '''New!''' Emotional Support Conversation Generation based on LLM (B/M/P)===
+Emotional support conversation aims to reduce individuals' emotional distress through social interaction and help them understand and cope with the challenges they face. Using LLM to provide emotional support is a promising technology which can be used in customer service chats, mental health support and so on. We need students for this topic. We expect you have a background in dialogue generation and programming skills in Python.
+Please contact Jing Li [jing.li@cs.uni-goettingen.de]
+===  * '''New!''' Rumor control and detection method for social networks based on GCN===
+In social networks, rumors spread quickly and have a wide impact. Through GCN, the complex relationships between nodes in the network can be effectively captured, and the detection and propagation paths of rumors can be modeled and controlled. The core of this method is to improve the ability to identify and control the spread of rumors by jointly learning user behavior, information content, and network topology by building an information propagation graph. We need students in this topic. We expect you have a background in rumor detection and programming skills in Python.
+Please contact Fei Gao [fei.gao@cs.uni-goettingen.de]
+===  [Occupied]  Image-to-Image Translation of Different Nightlight Image Types (B/M/P) ===
 Nightlight intensities have been proven to be a good indicator for socio-economic status. However, for long-term temporal analyses their use can be challenging, as different satellites for sensing nightlight intensities operated at different times (DMSP OLS 1992-2014 and VIIRS 2012-2023). Both types differ not only in resolution, but there is also a big discrepancy in the optical appearance and value ranges. To obtain consistent nightlight images for temporal analysis, Image-to-Image Translation methods shall be used in this project/thesis for the conversion between both types. Finally the performance of the translated and original nightlight images for a regression on socio-economic indicators shall be evaluated.
@@ Line 69: / Line 125: @@
 Please contact Dr. Tingting Yuan [tingting.yuan@cs.uni-goettingen.de]] (B/M/P)
-===   *  Privacy-preserved Video Analytics===
+===   * '''[Closed]''' Privacy-preserved Video Analytics===
 This project/thesis topic focuses on the protection of privacy in video analytics.
@@ Line 84: / Line 140: @@
 [1] Pecam: privacy-enhanced video streaming and analytics via securely-reversible transformation [https://dl.acm.org/doi/abs/10.1145/3447993.3448618].
-===  *  AI for networking adaption  ===
+===  * '''[Closed]''' AI for networking adaption  ===
 In this project/theses topic, you will explore how to make AI meets networking requirements (e.g., fluctuating network states).
 You will (1) deploy and test Genet[1]; (2)extend the Genet environment to multi-client environment (e.g., ABR); (3) deploy multi-agent algorithms on Genet and valid the performance.
@@ Line 109: / Line 165: @@
 Please contact Dr. Tingting Yuan [tingting.yuan@cs.uni-goettingen.de] (B/M/P)
-=== * '''New!''' Socioecomonic analysis based on spatiotemporal and linguistic analysis on microblogging data ===
+=== * '''[Closed]''' Socioecomonic analysis based on spatiotemporal and linguistic analysis on microblogging data ===
 Identifying the socioeconomic status (SES) of users in social media like Twitter or Weibo is useful e.g., for digitized advertisements and social policies. This study aims to collect profiles of Twitter users on selected topics such as culture or foreign language learning, extract the temporal, spatial and linguistic features, and compare different classification algorithms (e.g., decision tree, random forest, na\"{i}ve Bayes, deep learning, and Gaussian processes classifier) to predict the socioeconomic status.
@@ Line 118: / Line 174: @@
 Please contact  Prof. Xiaoming Fu [fu@cs.uni-goettingen.de](B/M/P)
 === [Closed] Super resolution technique for efficient video delivery ===
@@ Line 157: / Line 212: @@
 Please contact  Dr.Tingting Yuan [tingting.yuan@cs.uni-goettingen.de ] and Weijun Wang [weijun.wang@informatik.uni-goettingen.de](B/M/P)
--->
 == Ongoing Topics ==
@@ Line 186: / Line 238: @@
 | Completed by Jason
 |-
-| Video analytics with deep reinforcement learning (Master Thesis)
+| Neural video analytics(Master Thesis)
 |[http://www.net.informatik.uni-goettingen.de/?q=people/dr-tingting-yuan Tingting Yuan]
 |