Existing open-vocabulary image segmentation methods require a fine-tuning step on mask
labels and/or image-text datasets. Mask labels are labor-intensive which limits the number of …

The crux of learning vision-language models is to extract semantically aligned information
from visual and linguistic data. Existing attempts usually face the problem of coarse …

This paper addresses text-supervised semantic segmentation aiming to learn a model
capable of segmenting arbitrary visual concepts within images by using only image-text …

Contemporary cutting-edge open-vocabulary segmentation approaches commonly rely on
image-mask-text triplets, yet this restricted annotation is labour-intensive and encounters …

We present lazy visual grounding, a two-stage approach of unsupervised object mask
discovery followed by object grounding, for open-vocabulary semantic segmentation. Plenty …

A serious issue that harms the performance of zero-shot visual recognition is named
objective misalignment, ie, the learning objective prioritizes improving the recognition …

Vision-language models (VLMs) have demonstrated remarkable open-vocabulary object
recognition capabilities, motivating their adaptation for dense prediction tasks like …

The pre-trained vision-language model, exemplified by CLIP, advances zero-shot semantic
segmentation by aligning visual features with class embeddings through a transformer …

Despite the significant progress in deep learning for dense visual recognition problems,
such as semantic segmentation, traditional methods are constrained by fixed class sets …

In semantic segmentation, generalizing a visual system to both seen categories and novel
categories at inference time has always been practically valuable yet challenging. To enable …