# Deep Semantic Matching with Foreground Detection and Cycle-Consistency

arXiv: https://arxiv.org/abs/2004.00144 · HTML: https://mlsanity.com/paper/2004.00144/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 31 Mar 2020.

## Built on by (most cited first, 3 of 3)

- [Semantic-Aware Fine-Grained Correspondence](https://mlsanity.com/paper/2207.10456/) (2022)
- [Learning Contrastive Representation for Semantic Correspondence](https://mlsanity.com/paper/2109.10967/) (2021)
- [Semantic Correspondence: Unified Benchmarking and a Strong Baseline](https://mlsanity.com/paper/2505.18060/) (2025)

## Abstract (arXiv)

Establishing dense semantic correspondences between object instances remains a challenging problem due to background clutter, significant scale and pose differences, and large intra-class variations. In this paper, we address weakly supervised semantic matching based on a deep network where only image pairs without manual keypoint correspondence annotations are provided. To facilitate network training with this weaker form of supervision, we 1) explicitly estimate the foreground regions to suppress the effect of background clutter and 2) develop cycle-consistent losses to enforce the predicted transformations across multiple images to be geometrically plausible and consistent. We train the proposed model using the PF-PASCAL dataset and evaluate the performance on the PF-PASCAL, PF-WILLOW, and TSS datasets. Extensive experimental results show that the proposed approach performs favorably against the state-of-the-art methods.