Search papers, labs, and topics across Lattice.
The paper introduces the Plain Mask Transformer (PMT), a Transformer-based segmentation decoder designed to operate on top of frozen Vision Foundation Model (VFM) features for both image and video segmentation. PMT aims to preserve the benefits of encoder-only models (simplicity, low latency) while leveraging the multi-task capabilities of frozen VFMs. Experiments show that PMT achieves competitive accuracy with significantly reduced latency compared to existing frozen-encoder and finetuned methods on image and video segmentation benchmarks.
Get 3x faster image segmentation and comparable video segmentation performance to fine-tuned models, all while keeping your vision encoder frozen.
Vision Foundation Models (VFMs) pre-trained at scale enable a single frozen encoder to serve multiple downstream tasks simultaneously. Recent VFM-based encoder-only models for image and video segmentation, such as EoMT and VidEoMT, achieve competitive accuracy with remarkably low latency, yet they require finetuning the encoder, sacrificing the multi-task encoder sharing that makes VFMs practically attractive for large-scale deployment. To reconcile encoder-only simplicity and speed with frozen VFM features, we propose the Plain Mask Decoder (PMD), a fast Transformer-based segmentation decoder that operates on top of frozen VFM features. The resulting model, the Plain Mask Transformer (PMT), preserves the architectural simplicity and low latency of encoder-only designs while keeping the encoder representation unchanged and shareable. The design seamlessly applies to both image and video segmentation, inheriting the generality of the encoder-only framework. On standard image segmentation benchmarks, PMT matches the frozen-encoder state of the art while running up to ~3x faster. For video segmentation, it even performs on par with fully finetuned methods, while being up to 8x faster than state-of-the-art frozen-encoder models. Code: https://github.com/tue-mps/pmt.
