This paper presents a rule-based procedural generation approach for layout designby design knowledge encoding. Taking linear shopping centres as example, the proposedmethod encodes the layout design elements of the wa...This paper presents a rule-based procedural generation approach for layout designby design knowledge encoding. Taking linear shopping centres as example, the proposedmethod encodes the layout design elements of the walkway space, tenant areas, and staircases into generative rules based on geometric operations. The generative rules integratethe shopping centre layout’s spatial patterns and geometric features and incorporate userspecified interaction parameters to form a generation tool prototype for early-stage layoutdesign. The results show that the method can deal with the complex spatial elements of linearshopping centres and provide design references for architects, which helps combine generativealgorithms with the design process.展开更多
Underwater scene is one of the most marvelous environments in the world. In this study, we present an efficient procedural modeling and rendering system to generate marine ecosystems for swim-through graphic applicati...Underwater scene is one of the most marvelous environments in the world. In this study, we present an efficient procedural modeling and rendering system to generate marine ecosystems for swim-through graphic applications. To produce realistic and natural underwater scenes, several techniques and algorithms have been presented and introduced. First, to distribute sealife naturally on a seabed, we employ an ecosystem simulation that considers the influence of the underwater environment. Second, we propose a two-level procedural modeling system to generate sealife with unique biological features. At the base level, a series of grammars are designed to roughly represent underwater sealife on a central processing unit(CPU). Then at the fine level, additional details of the sealife are created and rendered using graphic processing units(GPUs). Such a hybrid CPU-GPU framework best adopts sequential and parallel computation in modeling a marine ecosystem, and achieves a high level of performance.Third, the proposed system integrates dynamic simulations in the proposed procedural modeling process to support dynamic interactions between sealife and the underwater environment, where interactions and physical factors of the environment are formulated into parameters and control the geometric generation at the fine level. Results demonstrate that this system is capable of generating and rendering scenes with massive corals and sealife in real time.展开更多
Online content generation enables automatic and adaptive creation of diverse and playable game content for maximizing player experience or testing Artificial Intelligence(Al)algorithms.Multiple diversity metrics have ...Online content generation enables automatic and adaptive creation of diverse and playable game content for maximizing player experience or testing Artificial Intelligence(Al)algorithms.Multiple diversity metrics have been formulated on different content facets in the literature,while some of them conflict with one another.Existing work addresses this multi-dimensional diversity nature by converting those metrics into one term that is further used to direct the training of content generators.However,each generator is trained to meet the preference specified by the weights and fails to fully interpret the relationships among these metrics or provide different trade-offs.This paper proposes a multi-objective procedural content generation via reinforcement learning to train a set of generators that create diverse game content in an online manner while balancing the trade-off between multiple diversity metrics with playability as a constraint.Our framework is compared with state-of-the-art approaches on the commonly used Mario-Al benchmark.Results show that our framework is capable of increasing the diversity of the generator distribution while accelerating the convergence during the early stages of model training.Our approach enables researchers,designers,and practitioners to gain a better understanding of the relationship among conflicting diversity metrics,allowing them to generate content more efficiently and accurately tailored to specific needs.展开更多
In this paper, we investigate a modified differential-difference KP equation which is shown to have a continuum limit into the m KP equation. It is also shown that the solution of the modified differential-difference ...In this paper, we investigate a modified differential-difference KP equation which is shown to have a continuum limit into the m KP equation. It is also shown that the solution of the modified differential-difference KP equation is related to the solution of the differential-difference KP equation through a Miura transformation. We first present the Grammian solution to the modified differential-difference KP equation, and then produce a coupled modified differential-difference KP system by applying the source generation procedure. The explicit N-soliton solution of the resulting coupled modified differential-difference system is expressed in compact forms by using the Grammian determinant and Casorati determinant. We also construct and solve another form of the self-consistent sources extension of the modified differential-difference KP equation, which constitutes a B?cklund transformation for the differentialdifference KP equation with self-consistent sources.展开更多
In this paper, we apply the source generation procedure to the coupled 2D Toda lattice equation （also called Pfaffianized 2D Toda lattice）, then we get a more generalized system which is the coupled 2D Toda lattice ...In this paper, we apply the source generation procedure to the coupled 2D Toda lattice equation （also called Pfaffianized 2D Toda lattice）, then we get a more generalized system which is the coupled 2D Toda lattice with self-consistent sources （p-2D TodaESCS）, and a pfaman type solution of the new system is given. Consequently, by using the reduction of the pfaffian solution to the determinant form, this new system can not only be reduced to the 2D TodaESCS, but be reduced to the coupled 2D Toda lattice equation. This result indicates that the p-2D TodaESCS is also a pfafilan version of the 2D TodaESCS, which implies the commutativity between the source generation procedure and Pfaffianization is valid to the semi-discrete soliton equation.展开更多
New type of variable-coefficient KP equation with self-consistent sources and its Grammian solutions are obtained by using the source generation procedure.
In game design,how to balance narrative coherence and procedural level generation has always been a difficult problem.This paper proposes a method that can automatically extract playable narrative units from story tex...In game design,how to balance narrative coherence and procedural level generation has always been a difficult problem.This paper proposes a method that can automatically extract playable narrative units from story texts and combine them with dynamically generated levels.This method relies on deep semantic parsing,structured segmentation and playability constraints to decompose narrative text into atomic fragments that can be transformed into interactive environments.The system design is divided into two stages:In the first stage,narrative extraction is completed through dependency relationship and discourse analysis;The second stage achieves adaptive level generation by means of constraint-driven grammar and reinforcement feedback.The experiment was based on 124,600 narrative samples from mythological,fantasy and contemporary interactive novels,generating 19,420 independent narrative units and 1,250 levels.Compared with the random concatenation and grammar branch methods,this method significantly improves in narrative coherence,diversity and playability.Specifically,the average coherence score reached 7.83±0.42,which was significantly higher than 6.11±0.51 of the Grammar-driven method.The narrative-mechanism correspondence index reached 0.74±0.03,exceeding the current benchmark.The overall results show that the combination of narrative extraction and procedural generation can not only maintain the integrity of the story but also provide a feasible direction for the flexibility and scalability of the game.展开更多
Autonomous Driving Systems(ADS)are safety-critical.Abundant and various driving scenarios are required to train accurate and robust models,and comprehensively test each module of autonomous driving systems(i.e.,percep...Autonomous Driving Systems(ADS)are safety-critical.Abundant and various driving scenarios are required to train accurate and robust models,and comprehensively test each module of autonomous driving systems(i.e.,perception,tracking,prediction,planning,and control modules).However,collecting driving scenario data from the real physical world is expensive and inefficient.Most existing works generate simulated driving scenarios by varying the behaviors of dynamic objects on simple road networks(e.g.,highways),while the influence of roadside structures and scenarios with complex road networks are not considered.This paper proposes a novel driving scenario generation approach,Automated Scenario Crafting(AutoSceCraft),to automatically produce abundant driving scenarios containing various road networks,traffic rules,roadside structures,and dynamic objects at low cost.To validate the effectiveness and efficiency of our proposed framework,AutoSceCraft is integrated into three popular driving simulators,including SMARTS,esmini,and CARLA.Numerical experiments and scenario visualization results show that AutoSceCraft can generate effectively and efficiently various driving scenarios from scratch for testing and training various modules(including perception,prediction,and planning modules)within autonomous driving systems.展开更多
基金funded by the National Natural Science Foundation of China(Grant No.52378008)the Postgraduate Research&Practice Innovation Program of Jiangsu Province(KYCX24_0418).
文摘This paper presents a rule-based procedural generation approach for layout designby design knowledge encoding. Taking linear shopping centres as example, the proposedmethod encodes the layout design elements of the walkway space, tenant areas, and staircases into generative rules based on geometric operations. The generative rules integratethe shopping centre layout’s spatial patterns and geometric features and incorporate userspecified interaction parameters to form a generation tool prototype for early-stage layoutdesign. The results show that the method can deal with the complex spatial elements of linearshopping centres and provide design references for architects, which helps combine generativealgorithms with the design process.
基金Project supported by the Zhejiang Provincial Natural Science Foundation of China(No.LY13F020002)the National Natural Science Foundation of China(No.61272301)+1 种基金the National Key Technology R&D Program of China(No.2012BAH35B03)the Fundamental Research Funds for the Central Universities,China
文摘Underwater scene is one of the most marvelous environments in the world. In this study, we present an efficient procedural modeling and rendering system to generate marine ecosystems for swim-through graphic applications. To produce realistic and natural underwater scenes, several techniques and algorithms have been presented and introduced. First, to distribute sealife naturally on a seabed, we employ an ecosystem simulation that considers the influence of the underwater environment. Second, we propose a two-level procedural modeling system to generate sealife with unique biological features. At the base level, a series of grammars are designed to roughly represent underwater sealife on a central processing unit(CPU). Then at the fine level, additional details of the sealife are created and rendered using graphic processing units(GPUs). Such a hybrid CPU-GPU framework best adopts sequential and parallel computation in modeling a marine ecosystem, and achieves a high level of performance.Third, the proposed system integrates dynamic simulations in the proposed procedural modeling process to support dynamic interactions between sealife and the underwater environment, where interactions and physical factors of the environment are formulated into parameters and control the geometric generation at the fine level. Results demonstrate that this system is capable of generating and rendering scenes with massive corals and sealife in real time.
基金supported by the National Key R&D Program of China(No.2023YFE0106300)the National Natural Science Foundation of China(Nos.62250710682 and 62476119)internal grant of the Lingnan University,Hong Kong,China.
文摘Online content generation enables automatic and adaptive creation of diverse and playable game content for maximizing player experience or testing Artificial Intelligence(Al)algorithms.Multiple diversity metrics have been formulated on different content facets in the literature,while some of them conflict with one another.Existing work addresses this multi-dimensional diversity nature by converting those metrics into one term that is further used to direct the training of content generators.However,each generator is trained to meet the preference specified by the weights and fails to fully interpret the relationships among these metrics or provide different trade-offs.This paper proposes a multi-objective procedural content generation via reinforcement learning to train a set of generators that create diverse game content in an online manner while balancing the trade-off between multiple diversity metrics with playability as a constraint.Our framework is compared with state-of-the-art approaches on the commonly used Mario-Al benchmark.Results show that our framework is capable of increasing the diversity of the generator distribution while accelerating the convergence during the early stages of model training.Our approach enables researchers,designers,and practitioners to gain a better understanding of the relationship among conflicting diversity metrics,allowing them to generate content more efficiently and accurately tailored to specific needs.
基金Supported by the National Natural Science Foundation of China under Grant Nos.11601247 and 11605096the Natural Science Foundation of Inner Mongolia Autonomous Region under Grant Nos.2016MS0115 and 2015MS0116the Innovation Fund Programme of Inner Mongolia University No.201611155
文摘In this paper, we investigate a modified differential-difference KP equation which is shown to have a continuum limit into the m KP equation. It is also shown that the solution of the modified differential-difference KP equation is related to the solution of the differential-difference KP equation through a Miura transformation. We first present the Grammian solution to the modified differential-difference KP equation, and then produce a coupled modified differential-difference KP system by applying the source generation procedure. The explicit N-soliton solution of the resulting coupled modified differential-difference system is expressed in compact forms by using the Grammian determinant and Casorati determinant. We also construct and solve another form of the self-consistent sources extension of the modified differential-difference KP equation, which constitutes a B?cklund transformation for the differentialdifference KP equation with self-consistent sources.
基金Supported by the Fundamental Research Funds for the Central Universitiesthe Research Funds of Renmin University of China under Grant No. 07XNA013
文摘In this paper, we apply the source generation procedure to the coupled 2D Toda lattice equation （also called Pfaffianized 2D Toda lattice）, then we get a more generalized system which is the coupled 2D Toda lattice with self-consistent sources （p-2D TodaESCS）, and a pfaman type solution of the new system is given. Consequently, by using the reduction of the pfaffian solution to the determinant form, this new system can not only be reduced to the 2D TodaESCS, but be reduced to the coupled 2D Toda lattice equation. This result indicates that the p-2D TodaESCS is also a pfafilan version of the 2D TodaESCS, which implies the commutativity between the source generation procedure and Pfaffianization is valid to the semi-discrete soliton equation.
基金Supported by the NSF of Henan Province(112300410109)Supported by the NSF of the Education Department(2010A110022)
文摘New type of variable-coefficient KP equation with self-consistent sources and its Grammian solutions are obtained by using the source generation procedure.
文摘In game design,how to balance narrative coherence and procedural level generation has always been a difficult problem.This paper proposes a method that can automatically extract playable narrative units from story texts and combine them with dynamically generated levels.This method relies on deep semantic parsing,structured segmentation and playability constraints to decompose narrative text into atomic fragments that can be transformed into interactive environments.The system design is divided into two stages:In the first stage,narrative extraction is completed through dependency relationship and discourse analysis;The second stage achieves adaptive level generation by means of constraint-driven grammar and reinforcement feedback.The experiment was based on 124,600 narrative samples from mythological,fantasy and contemporary interactive novels,generating 19,420 independent narrative units and 1,250 levels.Compared with the random concatenation and grammar branch methods,this method significantly improves in narrative coherence,diversity and playability.Specifically,the average coherence score reached 7.83±0.42,which was significantly higher than 6.11±0.51 of the Grammar-driven method.The narrative-mechanism correspondence index reached 0.74±0.03,exceeding the current benchmark.The overall results show that the combination of narrative extraction and procedural generation can not only maintain the integrity of the story but also provide a feasible direction for the flexibility and scalability of the game.
基金supported by the National Key R&D Program of China(No.2023YFE0106300)the National Natural Science Foundation of China(Nos.62476119 and 62250710682)the Guangdong Major Project of Basic and Applied Basic Research(No.2023B0303000010).
文摘Autonomous Driving Systems(ADS)are safety-critical.Abundant and various driving scenarios are required to train accurate and robust models,and comprehensively test each module of autonomous driving systems(i.e.,perception,tracking,prediction,planning,and control modules).However,collecting driving scenario data from the real physical world is expensive and inefficient.Most existing works generate simulated driving scenarios by varying the behaviors of dynamic objects on simple road networks(e.g.,highways),while the influence of roadside structures and scenarios with complex road networks are not considered.This paper proposes a novel driving scenario generation approach,Automated Scenario Crafting(AutoSceCraft),to automatically produce abundant driving scenarios containing various road networks,traffic rules,roadside structures,and dynamic objects at low cost.To validate the effectiveness and efficiency of our proposed framework,AutoSceCraft is integrated into three popular driving simulators,including SMARTS,esmini,and CARLA.Numerical experiments and scenario visualization results show that AutoSceCraft can generate effectively and efficiently various driving scenarios from scratch for testing and training various modules(including perception,prediction,and planning modules)within autonomous driving systems.
