Crowd Counting and Density Estimation by Trellis Encoder-Decoder Network

TL;DR

This paper introduces TEDnet, a novel trellis encoder-decoder network for crowd counting that improves density map quality and achieves state-of-the-art results by using multi-path decoding, dense skip connections, and a new loss function.

Contribution

The paper proposes a trellis architecture with multiple decoding paths, dense skip connections, and a combinatorial loss for improved crowd density estimation.

Findings

Achieves up to 14% MAE improvement on benchmarks.

Outperforms previous methods in density map quality.

Enhances convergence speed with the new loss function.

Abstract

Crowd counting has recently attracted increasing interest in computer vision but remains a challenging problem. In this paper, we propose a trellis encoder-decoder network (TEDnet) for crowd counting, which focuses on generating high-quality density estimation maps. The major contributions are four-fold. First, we develop a new trellis architecture that incorporates multiple decoding paths to hierarchically aggregate features at different encoding stages, which can handle large variations of objects. Second, we design dense skip connections interleaved across paths to facilitate sufficient multi-scale feature fusions and to absorb the supervision information. Third, we propose a new combinatorial loss to enforce local coherence and spatial correlation in density maps. By distributedly imposing this combinatorial loss on intermediate outputs, gradient vanishing can be largely alleviated for better back-propagation and faster convergence. Finally, our TEDnet achieves new state-of-the art performance on four benchmarks, with an improvement up to 14% in terms of MAE.