Hypercolumns for Object Segmentation and Fine-grained Localization

TL;DR

This paper introduces hypercolumns, a method combining features from multiple CNN layers to improve fine-grained localization tasks like detection, segmentation, and keypoint localization, achieving significant performance gains.

Contribution

The paper proposes hypercolumns as a novel pixel descriptor combining multi-layer CNN features, enhancing localization accuracy across various tasks.

Findings

Improved mean AP^r from 49.7 to 60.0 in detection and segmentation.

Achieved a 3.3 point boost in keypoint localization.

Gained a 6.6 point improvement in part labeling.

Abstract

Recognition algorithms based on convolutional networks (CNNs) typically use the output of the last layer as feature representation. However, the information in this layer may be too coarse to allow precise localization. On the contrary, earlier layers may be precise in localization but will not capture semantics. To get the best of both worlds, we define the hypercolumn at a pixel as the vector of activations of all CNN units above that pixel. Using hypercolumns as pixel descriptors, we show results on three fine-grained localization tasks: simultaneous detection and segmentation[22], where we improve state-of-the-art from 49.7[22] mean AP^r to 60.0, keypoint localization, where we get a 3.3 point boost over[20] and part labeling, where we show a 6.6 point gain over a strong baseline.