Cosine meets Softmax: A tough-to-beat baseline for visual grounding

TL;DR

This paper introduces a simple yet effective baseline for visual grounding in autonomous driving, outperforming previous methods by leveraging cosine similarity and minimal design, challenging the need for complex models.

Contribution

The authors propose a minimalistic approach using cosine distance and pre-trained embeddings, achieving state-of-the-art results with less complexity.

Findings

Achieved 68.7% AP50 accuracy on Talk2Car dataset

Outperformed previous state-of-the-art by 8.6%

Showed simpler methods can be competitive with complex models

Abstract

In this paper, we present a simple baseline for visual grounding for autonomous driving which outperforms the state of the art methods, while retaining minimal design choices. Our framework minimizes the cross-entropy loss over the cosine distance between multiple image ROI features with a text embedding (representing the give sentence/phrase). We use pre-trained networks for obtaining the initial embeddings and learn a transformation layer on top of the text embedding. We perform experiments on the Talk2Car dataset and achieve 68.7% AP50 accuracy, improving upon the previous state of the art by 8.6%. Our investigation suggests reconsideration towards more approaches employing sophisticated attention mechanisms or multi-stage reasoning or complex metric learning loss functions by showing promise in simpler alternatives.