This book and its companion volume, LNCS vols. 7331 and 7332, constitute the proceedings of the Third International Conference on Swarm Intelligence, ICSI 2012, held in Shenzhen, China in June 2012. The 145 revised fu...
详细信息
ISBN:
(数字)9783642309762
ISBN:
(纸本)9783642309755
This book and its companion volume, LNCS vols. 7331 and 7332, constitute the proceedings of the Third International Conference on Swarm Intelligence, ICSI 2012, held in Shenzhen, China in June 2012. The 145 revised full papers presented were carefully reviewed and selected from 247 submissions. The papers are organized in 27 cohesive sections covering all major topics of swarm intelligence research and developments.
The 2010 Pacific-Rim Conference on Multimedia (PCM 2010) was held in Shanghai at Fudan University, during September 21–24, 2010. Since its inauguration in 2000, PCM has been held in various places around the Pacific ...
详细信息
ISBN:
(数字)9783642156960
ISBN:
(纸本)9783642156953
The 2010 Pacific-Rim Conference on Multimedia (PCM 2010) was held in Shanghai at Fudan University, during September 21–24, 2010. Since its inauguration in 2000, PCM has been held in various places around the Pacific Rim, namely Sydney (PCM 2000), Beijing (PCM 2001), Hsinchu (PCM 2002), Singapore (PCM 2003), Tokyo (PCM 2004), Jeju (PCM 2005), Zhejiang (PCM 2006), Hong Kong (PCM 2007), Tainan (PCM 2008), and Bangkok (PCM 2009). PCM is a major annual international conference organized as a forum for the dissemination of state-of-the-art technological advances and research results in the fields of theoretical, experimental, and applied multimedia analysis and processing. PCM 2010 featured a comprehensive technical program which included 75 oral and 56 poster presentations selected from 261 submissions from Australia, Canada, China, France, Germany, Hong Kong, India, Iran, Italy, Japan, Korea, Myanmar, Norway, Singapore, Taiwan, Thailand, the UK, and the USA. Three distinguished researchers, Prof. Zhi-Hua Zhou from Nanjing University, Dr. Yong Rui from Microsoft, and Dr. Tie-Yan Liu from Microsoft Research Asia delivered three keynote talks to the conference. We are very grateful to the many people who helped to make this conference a s- cess. We would like to especially thank Hong Lu for local organization, Qi Zhang for handling the publication of the proceedings, and Cheng Jin for looking after the c- ference website and publicity. We thank Fei Wu for organizing the special session on large-scale multimedia search in the social network settings.
Iterative inference approaches have shown promising success in the task of multi-view depth estimation. However, these methods put excessive emphasis on the universal inter-view correspondences while neglecting the co...
详细信息
Iterative inference approaches have shown promising success in the task of multi-view depth estimation. However, these methods put excessive emphasis on the universal inter-view correspondences while neglecting the correspondence ambiguity in regions of low texture and depth discontinuous areas. Thus, they are prone to produce inaccurate or even erroneous depth estimations, which is further exacerbated cumulative errors especially in the iterative pipeline, providing unreliable information in many real-world scenarios. In this paper, we revisit this issue from the intra-view Contextual Hints and introduce a novel enhancing iterative approach, named EnIter. Concretely, at the beginning of each iteration, we present a Depth Intercept (DI) modulator to provide more accurate depth by aggregating neighbor uncertainty, correlation volume of reference and normal. This plug and play modulator is effective at intercepting the erroneous depth estimations with implicit guidance from the universal correlation contextual hints, especially for the challenging regions. Furthermore, at the end of each iteration, we refine the depth map with another plug and play modulator termed as Depth Refine (DR). It mines the latent structure knowledge of reference Contextual Hints and establishes one-way dependency using local attention from reference features to depth, yielding delicate depth in details. Extensive experiment demonstrates that our method not only achieves state-of-the-art performance over existing models but also exhibits remarkable universality in popular iterative pipelines, e.g., CasMVS, UCSNet, TransMVS, UniMVS.
We systematically investigate the effects of interlayer, inter-ring dipolar coupling and the number of bilayer repeats on the magnetization reversal process of [Co/Pd]/Au(t)/[Co/Pd] pseudo-spin-valve (PSV) nano-rings....
详细信息
After two decades, data processing has finally, and probably forever, found its niche among civil engineering and construction (CEC) professionnals, through word processors, digitizing tables, management software, and...
详细信息
ISBN:
(数字)9781468474046
ISBN:
(纸本)9781850912538
After two decades, data processing has finally, and probably forever, found its niche among civil engineering and construction (CEC) professionnals, through word processors, digitizing tables, management software, and increasingly via drawing software and computer-aided design (CAD), recently, robots have even started invading work sites. What are the main trends of CAD and robotics in the field of architecture and civil enginee ring? What type of R&D effort do university and industrial laboratories undertake to devise the professional software that will be on the market in the next three to five years? These are the issues which will be addressed during this symposium. To this effect, we have planned concurrently an equipment and software show, as well as a twofold conference. Robotic is just starting in the field of civil engineering and construction. A pioneer, the Civil engineering Departement of Carnegie-Mellon University, in the United States, organized the first two international symposia, in 1984 and 1985 in Pittsburgh. This is the third meeting on the subject (this year, however, we have also included CAD). It constitutes the first large international symposium where CAD experts, specialists in architecture and CEC robotics will meet. From this standpoint, it should be an ideal forum for exchanging views and expe riences on a wide range of topics, and we hope it will give rise to novel applications and new syntheses. This symposium is intented for scientists, teachers, students and also for manufacturers and all CEC professionals.
When intelligent agents act in a stochastic environment, the principle of maximizing expected rewards is used to optimize their policies. The rationality of the maximum rewards becomes a single objective when agents’...
详细信息
When intelligent agents act in a stochastic environment, the principle of maximizing expected rewards is used to optimize their policies. The rationality of the maximum rewards becomes a single objective when agents’ decision problems are solved in most cases. This sometimes leads to the agents’ behaviors (the optimal policies for solving the decision problems) that are not legible. In other words, it is difficult for users (or other agents and even humans) to understand the agents’ intentions when they are executing the optimal policies. Hence, it becomes pertinent to consider the legibility of agents’ decision problems. The key challenge lies in formulating a proper legibility function in the problems. Using domain experts’ inputs leans to be subjective and inconsistent in specifying legibility values, and the manual approach quickly becomes infeasible in a complex problem domain. In this article, we aim to learn such a legibility function parallel to developing a (conventional) reward function. We adopt inverse reinforcement learning techniques to automate a legibility function in agents’ decision problems. We first demonstrate the effectiveness of the inverse reinforcement learning technique when legibility is solely considered in a decision problem. Things become complicated when both the reward and legibility functions are to be found. We develop a multi-objective inverse reinforcement learning method to automate the two functions in a good balance simultaneously. We vary problem domains in the performance study and provide empirical results in support.
暂无评论