| 1. |
- Goolaup, Sandhiya, 1985, et al.
(författare)
-
Developing a Theory of Surprise from Travelers’ Extraordinary Food Experiences
- 2018
-
Ingår i: Journal of Travel Research. - : SAGE Publications. - 0047-2875 .- 1552-6763. ; 57:2, s. 218-231
-
Tidskriftsartikel (refereegranskat)abstract
- The purpose of this research is to explore the extraordinary experiences of food tourists and to develop a theory of surprise in relation to a typology of food cultural capital. We draw on phenomenological interviews with 16 food tourists. We found that food tourists experienced surprise in different ways, depending on their food cultural capital. Food tourists who possessed a high level of cultural capital were surprised by the simplicity or complexity of the experience while those possessing a low level of cultural capital were surprised by the genuinity of the experience. Thus, we make an important theoretical contribution here as we learn that the resources food tourists possessed in the form of cultural capital conditioned the ways in which they conceived an extraordinary experience. More so, using the cultural capital perspective, we have also demonstrated the role of social context in contributing to creating an extraordinary experience.
|
|
| 2. |
|
|
| 3. |
- Rasheed, Hanoona, et al.
(författare)
-
Fine-tuned CLIP Models are Efficient Video Learners
- 2023
-
Ingår i: 2023 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR. - : IEEE COMPUTER SOC. - 9798350301298 - 9798350301304 ; , s. 6545-6554
-
Konferensbidrag (refereegranskat)abstract
- Large-scale multi-modal training with image-text pairs imparts strong generalization to CLIP model. Since training on a similar scale for videos is infeasible, recent approaches focus on the effective transfer of image-based CLIP to the video domain. In this pursuit, new parametric modules are added to learn temporal information and inter-frame relationships which require meticulous design efforts. Furthermore, when the resulting models are learned on videos, they tend to overfit on the given task distribution and lack in generalization aspect. This begs the following question: How to effectively transfer image-level CLIP representations to videos? In this work, we show that a simple Video Fine-tuned CLIP (ViFi-CLIP) baseline is generally sufficient to bridge the domain gap from images to videos. Our qualitative analysis illustrates that the framelevel processing from CLIP image-encoder followed by feature pooling and similarity matching with corresponding text embeddings helps in implicitly modeling the temporal cues within ViFi-CLIP. Such fine-tuning helps the model to focus on scene dynamics, moving objects and inter-object relationships. For low-data regimes where full fine-tuning is not viable, we propose a bridge and prompt approach that first uses fine-tuning to bridge the domain gap and then learns prompts on language and vision side to adapt CLIP representations. We extensively evaluate this simple yet strong baseline on zero-shot, base-to-novel generalization, few-shot and fully supervised settings across five video benchmarks. Our code and pre-trained models are available at https://github.com/muzairkhattak/ViFi-CLIP.
|
|
| 4. |
- Rasheed, Muhammad Babar, et al.
(författare)
-
Delay and energy consumption analysis of priority guaranteed MAC protocol for wireless body area networks
- 2017
-
Ingår i: Wireless networks. - : Springer. - 1022-0038 .- 1572-8196. ; 23:4, s. 1249-1266
-
Tidskriftsartikel (refereegranskat)abstract
- Wireless body area networks are captivating growing interest because of their suitability for wide range of applications. However, network lifetime is one of the most prominent barriers in deploying these networks for most applications. Moreover, most of these applications have stringent QoS requirements such as delay and throughput. In this paper, the modified superframe structure of IEEE 802.15.4 based MAC protocol is proposed which addresses the aforementioned problems and improves the energy consumption efficiency. Moreover, priority guaranteed CSMA/CA mechanism is used where different priorities are assigned to body nodes by adjusting the data type and size. In order to save energy, a wake-up radio based mechanism to control sleep and active modes of body sensors are used. Furthermore, a discrete time finite state Markov model to find the node states is used. Analytical expressions are derived to model and analyze the behavior of average energy consumption, throughput, packet drop probability, and average delay during normal and emergency data. Extensive simulations are conducted for analysis and validation of the proposed mechanism. Results show that the average energy consumption and delay are relatively higher during emergency data transmission with acknowledgment mode due to data collision and retransmission.
|
|
