以由下而上的搜尋為基礎的物件位置評估

Bottom-up Search for Top-down Objectness Estimation

王聰超(Cong-Chao Wang)

指導教授 ： 賴尚宏

國立清華大學/電機資訊學院/資訊工程學系/碩士(2014年)

https://doi.org/10.6843/NTHU.2014.00482

摘要

在本文中，我們提出一種與物件類別無關的物件可能性估計方法，能夠找到一副圖像中可能存在的物件的位置。為了實現這個目的，我們結合超級畫素與基於圖的分割兩種方法，高效地產生候選的一個物件區域集合，輔助後續物件偵測的工作。不僅使用分割演算法，我們還允許所有相鄰的超級畫素依據它們的相似程度彼此融合。然後，我們使用歸一化的邊界訊息來描述每一個物件位置的“明晰邊界”屬性，並利用一個提前訓練好的分類器來衡量這些位置的物件可能性。為了
發揮了窮舉搜索的優勢，同時又避免產生過多的重合度很高的物件位置，我們對我們方法的每一個步驟做了多樣化的操作，並使用非最大化抑制的方法減少產生的物件位置數量。完成這些工作後，我們能夠得到一個小且精的物件位置集合（在PASCAL VOC 2007的測試集上，實現77.4%的平均最優重合度（MABO）及94.4%召回率（DR））。通過結合不同的參數得到的結果，我們的方法的表現可以進一步提升，達到89.2%的MABO值和99.4%的DR 值。

關鍵字

自底向上的搜尋 ； 物件可能性估計 ； 物件偵測

並列摘要

In this thesis, we present a class-independent objectness estimation method searching possible object locations in one image. Towards this goal, we combine superpixels with graph-based segmentation to efficiently generate candidate regions for object detection. Beyond segmentation, our approach allows each of the hypotheses to iteratively merge with its neighboring superpixels based on their similarity. Then we use normed edge feature %which successfully reduces the noises of complex background's texture,to describe the close-boundary characteristics of the bounding box for each hypothesis and measure the associated objectness afterward by a pre-trained classifier. To utilize the advantage of exhaustive search and avoid generating too many high-overlap hypotheses, we diversify each step of our approach and apply non-maximal suppression(NMS) to refine the hypotheses. Thereafter, by using the proposed algorithm, we obtain a small set of high-quality hypothesized object locations(77.4% Mean Average Best Overlap(MABO) and 94.4% detection rate(DR) for all the objects in PASCAL VOC 2007 test set). By using the proposed strategies, the performance is increased to 89.2% MABO and 99.4% DR.

並列關鍵字

Bottom-up Search ； Objectness Estimation ； Object Detection

參考文獻

Achanta, R., Shaji, A., Smith, K., Lucchi, A., Fua, P., & Susstrunk, S. (2012). Slic superpixels compared to state-of-the-art superpixel methods. Pattern Analysis and

Alexe, B., Deselaers, T., & Ferrari, V. (2012). Measuring the objectness of image windows. Pattern Analysis and Machine Intelligence, IEEE Transactions on, 34(11),

2189–2202.

Branson, S., Beijbom, O., & Belongie, S. (2013). Efficient large-scale structured learning.

Bruce, N., & Tsotsos, J. (2006). Saliency based on information maximization. Advances