Straight to Shapes: Real-Time Detection of Encoded Shapes

Abstract

Current object detection approaches predict bounding boxes that provide little instance-specific information beyond location, scale and aspect ratio. In this work, we propose to regress directly to objects' shapes in addition to their bounding boxes and categories. It is crucial to find an appropriate shape representation that is compact and decodable, and in which objects can be compared for higherorder concepts such as view similarity, pose variation and occlusion. To achieve this, we use a denoising convolutional auto-encoder to learn a low-dimensional shape embedding space. We place the decoder network after a fast end-toend deep convolutional network that is trained to regress directly to the shape vectors provided by the auto-encoder. This yields what to the best of our knowledge is the first real-time shape prediction network, running at 35 FPS on a high-end desktop. With higher-order shape reasoning wellintegrated into the network pipeline, the network shows the useful practical quality of generalising to unseen categories that are similar to the ones in the training set, something that most existing approaches fail to handle.

Cell Detection with Star-Convex Polygons

“…To that end, we use star-convex polygons that we find well-suited to approximate the typically roundish shapes of cell nuclei in microscopy images. While Jetley et al [7] already investigated star-convex polygons for object detection in natural images, they found them to be inferior to more suitable shape representations for typical object classes in natural images, like people or bicycles.…”

Zero-Shot Object Detection: Learning to Simultaneously Recognize and Localize Novel Concepts

“…Another approach aiming at zero-shot segmentation is to learn a shape space shared with the novel objects. This technique, however, can only segment new object shapes that are very similar to the training set [16]. Along the second theme, some efforts have more recently been reported for object localization and tracking using natural language descriptions [14,26].…”

Explicit Shape Encoding for Real-Time Instance Segmentation

“…Explicit v.s Implicit Shape Representation A previous work with similar ideology has been done by Jetley et al [16]. They took the implicit shape representation path by first training an autoencoder on object binary mask.…”

“…We first compare the explicit shape encoding with the implicit shape encoding. As the previous work [16] provides a baseline for implicit shape representation with YOLO [30] as the base detector, to be fairly compared, we also trained the ESE-Seg with YOLO base detector, the dimension of the shape vector is also the same. We denote the model as "YOLO-Cheby (50)" and "YOLO-Cheby (20)".…”