光線追蹤技術曾經受限於龐大的運算需求，如今在系統運算能力快速的增強之下獲得改善，並讓光線追蹤重新獲得大量的關注。
本論文提出一個高效率的光線追蹤機制，透過尋找連續畫面間的一致性來呈現動態場景，對於許多常見的動態場景而言，從兩個連續畫面中的相同像素所發射出的光線，有很高的機率命中同一個物件。本論文採用額外的命中緩衝器來儲存當前場景中每條光線所相交的三角形之編號。為了呈現接續的場景，光線對樹狀結構做一般的走訪之前，每條光線主要的走訪機制為先將光線與前一個畫面所命中的三角形做相交測試，一旦光線命中同一個三角形，他們所相交的深度可以運用在樹的走訪中，排除掉許多需要做測試運算的節點。另外，假設在連續的畫面之間僅有少量的物體在場景中有移動，此時的
bounding volume hierarchies (BVH)樹狀結構可以更新前一個場景的樹狀結構來獲得，所以我們可以透過標記一個旗標來表示每一個節點是否被修改過。對這些相交到與當前場景相同三角形的光線，在走訪樹狀結構時，未被修改的節點可以直接的忽略。
對於我們自製的測試場景，有變動的三角形數量小於整體三角形數量的5%，實驗的結果顯示，光線與節點及光線與三角形的相交測試運算量上有超過一半的節省比例，這也導致超過50%的節省比例在軟體中光線追蹤的時間。對於這個常見的測試場景TOASTERS，在連續的畫面之間超過90%的三角形有移動，本方法仍可節省光線追蹤時間到達30%。

Once restricted by its enormous computational demand, ray-tracing has regained a great deal of attention nowadays due to the fast increase of systems’ computing power. This thesis proposes an efficient ray-tracing scheme to render dynamic scenes by exploring the coherence between consecutive frames. Since for many ordinary dynamic scenes, the ray emitted from the same pixels of two consecutive frames may hit the same object with high possibility. This thesis adopts an extra hit-buffer to store the index of primitive each ray intersects for the current scene. To render the following scene, for the trace of each ray the proposed scheme will first apply the ray-triangle intersection operation to the triangle it hits in the previous frame before proceeding to the normal tree traversal procedure. Once the ray hits the same triangle, the intersection depth can be used to exclude the test operation of many nodes during the tree traversal. In addition, by assuming that only few objects in a scene may move from frame to frame, the bounding volume hierarchies (BVH) tree structure of the current scene can be obtained by updating the tree structure of the previous one. Therefore, we can annotate each node of the tree by an additional flag to denote whether it has been modified or not. The unmodified nodes can be directly ignored during the tree traversal for those rays which hit the same primitives in the current scene. For our own test scenes whose total number of modified triangles is less than 5% of the overall triangle count, our experimental results show that more than one half of ray-node tests and ray-triangle intersection operations can be saved, which can result in more than 50% reduction in the software ray-tracing time. For the common test scene TOASTERS where more than 90% of triangles have been moved in consecutive scenes, the proposed scheme can still achieve the saving of trace time up to 30%.

論文審定書 i
摘要 ii
Abstract iii
目錄 iv
圖次 v
表次 viii
Chapter 1. 概論 1
1.1. 研究動機 1
1.2. 論文大綱 1
Chapter 2. 研究背景與相關研究 3
2.1. 光線追蹤介紹 3
2.2. Bounding Volume Hierarchies(BVH)介紹 5
2.3. 光線群組追蹤介紹(Packet Ray) 7
2.4. Frustum-AABB/OBB Intersection 10
2.5. BVH Traversal介紹 11
2.6. Ray-Triangle Intersection介紹 13
2.7. BVH Update 14
2.8. Object space temporal coherence 15
2.9. Image space temporal coherence 16
Chapter 3. 光線加速走訪連續性場景 18
3.1. 距離測試法 18
3.2. Update BVH Tree 20
3.3. 光線走訪更新後的BVH Tree 21
3.3.1. 主要光線走訪 21
3.3.2. 陰影光線走訪 24
3.3.3. 反射光線走訪 26
Chapter 4. 實驗結果 27
Chapter 5. 硬體設計架構之構想與探討 34
5.1.1. 硬體雛型架構與各模組介紹 34
5.1.2. 匯流排傳輸資料量之探討 37
5.1.3. 儲存空間之探討 39
Chapter 6. 結論與未來展望 40
參考文獻 43

[1] De Greve, “Reflections and Refractions in Ray Tracing,” http://www.flipcode.com/archives/reflection_transmission.pdf
[2] “Bounding volume hierarchies” http://en.wikipedia.org/wiki/Bounding_volume_hierarchy
[3] “K-d Tree” http://en.wikipedia.org/wiki/K-d_tree
[4] 吳俊霖, “階層式走訪之光線追蹤電路設計,”國立中山大學碩士論文, September.2015
[5] M. Hapala, T. Davidovič, I. Wald, V. Havran, and P. Slusallek. “Efficient stack-less BVH traversal for ray tracing,” in Proc. of the 27th Spring Conference on Computer Graphics , ACM, New York, NY, USA, pp. 7-12. 2011.
[6] S. Boulos, D. Edwards, J D. Lacewell, J. Kniss, J. Kautz, P. Shirley, and I. Wald. “Packet-based whitted and distribution ray tracing,” in Proc. of Graphics Interface, ACM, New York, NY, USA, pp. 177-184, 2007.
[7] J. Gunther, S. Popov, H.-P. Seidel, P. Slusallek, “Realtime ray tracing on GPU with BVH-based packet traversal,” IEEE Symposium on Interactive Ray Tracing, pp.113-118, Sep. 2007.
[8] Overbeck, R., Ramamoorthi, R., and Mark, W. R. 2008. “Large Ray Packets for Real-time Whitted Ray Tracing.” In IEEE/Eurographics Symposium on Interactive Ray Tracing 2008.
[9] U. Assarsson and T. Moller, “Optimized view frustum culling algorithms for bounding boxes,” Journal of Graphics Tools, vol. 5, pp. 9–22, 2000.
[10] A. Williams, S. Barrus, R.K. Morley, P. Shirley, “An efficient and robust ray–box intersection algorithm,” in Journal of Graphics Tools, pp.49-54, Jan. 2005.
[11] T. Möller, B.Trumbore, “Fast, minimum storage ray-triangle intersection,”in Journal of Graphics Tools, pp.21-28, Oct. 1997.
[12] I. Wald, “Realtime Ray Tracing and Interactive Global Illumination,” PhD dissertation, Saarland Univ., 2004.
[13] M. Shevtsov, A. Soupikov, and A. Kapustin, “Ray-Triangle Intersection Algorithm for Modern CPU Architectures,” Proc GraphiCon ’07, pp. 33-39, 2007.
[14] Jiri Havel , Adam Herout, “Yet Faster Ray-Triangle Intersection (Using SSE4)”, IEEE Transactions on Visualization and Computer Graphics, v.16 n.3, p.434-438, May 2010.
[15] KENSLER A., SHIRLEY P.: “Optimizing Ray-Triangle Intersection via Automated Search.” In Proceedings of the 2006 IEEE Symposium on Interactive Ray Tracing (Sept. 2006), pp. 33–38.
[16] Daniel Kopta , Thiago Ize , Josef Spjut , Erik Brunvand , Al Davis , Andrew Kensler, “Fast, effective BVH updates for animated scenes,” Proceedings of the ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games, March 09-11, 2012, Costa Mesa, California.
[17] Ingo Wald , Solomon Boulos , Peter Shirley, “Ray tracing deformable scenes using dynamic bounding volume hierarchies”, ACM Transactions on Graphics (TOG), v.26 n.1, p.6-es, January 2007.
[18] Lauterbach, C., Yoon, S.-E., Tuft, D., and Manocha, D. 2006. “RT-DEFORM: Interactive ray tracing of dynamic scenes using BVHs.” In Proceedings of the 2006 IEEE Symposium on Interactive Ray Tracing. 39--45.
[19] .Timothy A. Davis , Edward W. Davis, “Exploiting frame coherence with the temporal depth buffer in a distributed computing environment”, Proceedings of the 1999 IEEE symposium on Parallel visualization and graphics, p.29-38, October 25-26, 1999, San Francisco, California, USA.

[20] 3D-Digital Differential Analyzer(3D-DDA) algorithm
https://www.researchgate.net/figure/233899848_fig3_A-3D-DDA-ray-traversal- example-in-the-2D-space-The-3D-DDA-algorithm-checks-the-two
[21] J. Chapman, T. W. Calvert, and .T. Dill, “Exploiting temporal coherence in ray tracing,” Proceedings of Graphics Interface '90, pp. 196-204, 1990.
[22] 林弘, “單一及群組光線混合追蹤之光線追蹤電路設計,”國立中山大學碩士論文,Jul.2014