深度学习在计算机视觉中的应用-优快云博客

本文链接：https://blog.youkuaiyun.com/xw2017/article/details/86589227

wipe out ：消灭、彻底摧毁
E.g:The first convolution layer in ImageNet classification networks has a stride of 2 followed by a max pooling layer of stride 2x2, which can potentially wipe out most of the image signal present in a small object.
优秀短语：Organised annually from 2005 to present
E.g:Organised annually from 2005 to present, the challenge and its associated dataset has become accepted as the benchmark for object detection.

E.g:RPN assigns an anchor as positive only if overlap with a ground
truth bounding box is greater than 0.7.

bells and whistles：花里胡哨的东西
E.g:The results in Table 7 shows that our TridentNet with single-scale image as input could achieve 42.7 AP on testdev set without bells and whistles.

E.g：It demonstrates the effectiveness of our scale specific feature maps, which are generated by TridentNet with the same representation power.

热红外数据的优势：The thermal sensors have many advantages over others, such as the long-range imaging ability, the insensitivity to lighting conditions and the strong ability to penetrate haze and smog.

thermal crossover：热交叉
热红外数据的缺点：Thermal spectrum is disturbed by glass and thermal crossover.
be identical for :两个相同
E.g：These steps must be identical for both training and inference.

E.g：Once you understand how training works, understanding inference is a lot easier as it simply uses a subset of the steps involved in training.

off-the-shelf :现成的
E.g: Pose estimation is obtained by a pretrained off-the-shelf model.
progressively:逐渐的
E.g：The encoder-decoder network progressively down-samples img_a to a bottleneck layer, and then reverse the process to generate img_b

E.g:Over recent years, the landscape of computer vision has been drastically altered and pushed forward through the adoption of a fast, scalable, end-to-end learning framework, the Convolutional Neural Network (CNN).

has been drastically altered and pushed forward: 快速发展

E.g:In addition, we provide a comprehensive analysis of several state-of-the-art segmentation approaches using three complementary metrics that measure the spatial extent of the segmentation, the accuracy of the silhouette contours and the temporal coherence.

sihouette contours:（物体的）轮廓

E.g:Each video is accompanied by densely annotated（每个视频都是像素级标注）, pixel-accurate and per-frame ground truth segmentation.

E.g:Video object segmentation has been studied extensively in the past decade due to its importance in understanding video spatial-temporal structures as well as its value in industrial applications. Recently, data-driven algorithms (e.g. deep learning) have become the dominant approach to computer vision problems and one of the most important keys to their successes is the availability of large-scale datasets. Unfortunately, there does not exist a large-scale dataset for video object segmentation. As a result, recent algorithms for video object segmentation have to train their models on image datasets which do not contain any temporal information or video-specific characteristics, and thus would settle for suboptimal solutions. Besides, previous benchmark does not differentiate the object categories used in training and testing, and therefore cannot evaluate the true generalization ability of algorithms on unseen categories.

E.g:While it delivered impressive results, the practical usability of such algorithms is limited.
E.g: We observe that the major difficulty of applying FCNs to instance segmentation is that the similar image appearance may require different predictions but FCNs struggle at achieving this

analogous 类似的
eliminate 消除
E.g: eliminate the need for
significantly 显著地，相当数量地
E.g: leading to significantly faster inference
mitigate 缓和 || with the price of 代价是
E.g: This issue may be mitigated by using rotated ROIs, but with the price of a more complex pipeline.