Related papers: PolyMaX: General Dense Prediction with Mask Transf…

Segmenting unknown or anomalous object instances is a critical task in autonomous driving applications, and it is approached traditionally as a per-pixel classification problem. However, reasoning individually about each pixel without…

Clustering is a fundamental machine learning task which has been widely studied in the literature. Classic clustering methods follow the assumption that data are represented as features in a vectorized form through various representation…

Cell-level dense prediction is central to computational pathology, but remains challenging due to fine-grained histological structures, strong domain shifts, and costly dense annotations. Existing ViT-based pathology foundation models rely…

Image segmentation is about grouping pixels with different semantics, e.g., category or instance membership, where each choice of semantics defines a task. While only the semantics of each task differ, current research focuses on designing…

Semantic segmentation, which aims to classify every pixel in an image, is a key task in machine perception, with many applications across robotics and autonomous driving. Due to the high dimensionality of this task, most existing approaches…

In this work, instead of directly predicting the pixel-level segmentation masks, the problem of referring image segmentation is formulated as sequential polygon generation, and the predicted polygons can be later converted into segmentation…

Domain adaptive panoptic segmentation promises to resolve the long tail of corner cases in natural scene understanding. Previous state of the art addresses this problem with cross-task consistency, careful system-level optimization and…

Deep convolutional neural networks require large amounts of labeled data samples. For many real-world applications, this is a major limitation which is commonly treated by augmentation methods. In this work, we address the problem of…

Medical image segmentation typically adopts a point-wise convolutional segmentation head to predict dense labels, where each output channel is heuristically tied to a specific class. This rigid design limits both feature sharing and…

Dense prediction tasks typically employ encoder-decoder architectures, but the prevalent convolutions in the decoder are not image-adaptive and can lead to boundary artifacts. Different generalized convolution operations have been…

Multimodal Large Language Models (MLLMs) have demonstrated exceptional capabilities in high-level visual understanding. However, extending these models to fine-grained dense prediction tasks, such as semantic segmentation and depth…

Object pose estimation is a key perceptual capability in robotics. We propose a fully-convolutional extension of the PoseCNN method, which densely predicts object translations and orientations. This has several advantages such as improving…

Instance segmentation is one of the fundamental vision tasks. Recently, fully convolutional instance segmentation methods have drawn much attention as they are often simpler and more efficient than two-stage approaches like Mask R-CNN. To…

This paper introduces a diffusion-based framework for universal image segmentation, making agnostic segmentation possible without depending on mask-based frameworks and instead predicting the full segmentation in a holistic manner. We…

Masked autoencoding has achieved great success for self-supervised learning in the image and language domains. However, mask based pretraining has yet to show benefits for point cloud understanding, likely due to standard backbones like…

We propose a simple, efficient, yet powerful framework for dense visual predictions based on the conditional diffusion pipeline. Our approach follows a "noise-to-map" generative paradigm for prediction by progressively removing noise from a…

Pathology foundation models (PFMs) have rapidly advanced and are becoming a common backbone for downstream clinical tasks, offering strong transferability across tissues and institutions. However, for dense prediction (e.g., segmentation),…

By measuring, modeling and interpreting cosmological datasets, one can place strong constraints on models of the Universe. Central to this effort are summary statistics such as power spectra and bispectra, which condense the…

Inverse protein folding generates valid amino acid sequences that can fold into a desired protein structure, with recent deep-learning advances showing strong potential and competitive performance. However, challenges remain, such as…

Human motion prediction is a classical problem in computer vision and computer graphics, which has a wide range of practical applications. Previous effects achieve great empirical performance based on an encoding-decoding style. The methods…