International Journal of Computer Vision (IJCV) details the science and engineering of this rapidly growing field. Regular articles present major technical advances of broad general interest. Survey articles offer critical reviews of the state of the art and/or tutorial presentations of pertinent topics.

Coverage includes:

- Mathematical, physical and computational aspects of computer vision: image formation, processing, analysis, and interpretation; machine learning techniques; statistical approaches; sensors.
- Applications: image-based rendering, computer graphics, robotics, photo interpretation, image retrieval, video analysis and annotation, multi-media, and more.
- Connections with human perception: computational and architectural aspects of human vision.

The journal also features book reviews, position papers, editorials by leading scientific figures, as well as additional on-line material, such as still images, video sequences, data sets, and software.

Please note: the median time indicated below is computed over all the submitted manuscripts including the ones that are not put into the review pipeline at the onset of the review process. The typical time to first decision for manuscripts is approximately 96 days.

Special Issue on Unconstrained 3D Reconstruction and Rendering
Submission Date: 2026-01-15

Guest editors

    Rama Chellappa, Johns Hopkins University, USA
    Rakesh (Teddy) Kumar, Center for Vision Technologies, SRI International, USA
    Cheng Peng, University of Virginia, USA
    Ravi Ramamoorthi, University of California - San Diego, USA
    Deva Ramanan, Carnegie Mellon University, USA

Introduction

Photorealistic scene reconstruction from 2D images is a long-standing, fundamental research area that spans computer vision, graphics, and photogrammetry. Many applications from this research, e.g., in autonomous navigation, augmented reality, smart cities, disaster relief planning, etc., can greatly benefit society. While scene reconstruction, e.g., based on Neural Radiance Field, has seen significant interest and progress, in-the-wild, at-scale photorealistic 3D/4D reconstruction and novel view synthesis is still a highly complex problem. Broadly speaking, scene reconstruction can be decomposed into camera calibration and dense reconstruction/view synthesis. From a camera calibration perspective, robust feature matching and calibration from diverse sensor types remain challenging. From a dense reconstruction and rendering perspective, choosing the correct 3D representation and addressing various multi-view inconsistencies are still open questions. Finally, the optimal way to incorporate prior knowledge, such as that from large-scale 3D foundation models and generative models for reconstruction and rendering, is of great interests.    

Aims & Scope

This special issue invites innovative research papers that aim to address these challenges and propose novel techniques for robust 3D reconstruction at scale. Potential topics of interest include, but are not limited to:   

    Feature matching under challenging conditions: This includes methods for improving feature matching, such as matching from large baseline, eliminating visual ambiguities, addressing issues in image rotation, specular surfaces, low-texture regions, etc.    
    Novel sensory model/multi-sensory calibration: This includes methods for improving camera calibration for infrared, multi-spectral, satellite, non-parametric or other types of single or a mixture of novel camera models.    
    In-the-wild 3D/4D reconstruction and novel view synthesis: This includes methods for 3D/4D reconstruction and novel view synthesis from images acquired at different times, under nonideal quality (blur, non-Lambertian, compression, over and under saturation etc.), non-ideal calibration, at scale, or on dynamic, indoor and outdoor scenes.  Images may be sparse, or acquired from ground, aerial, satellite platforms.  
    3D scene/novel view synthesis representation: This includes methods that improve upon 3D/4D representations for reconstruction and rendering, such as neural implicit representation, explicit representation e.g., points, splats, surfels, mesh, etc., or methods that introduce novel 3D/4D representations.    
    Prior-based reconstruction and novel view synthesis: This includes feedforward methods on monocular depth estimation, 3D foundation model, or generative-model-assisted novel view synthesis  using scene priors that can improve reconstruction and rendering in-the-wild with sparse set of images and also at-scale. 

Moreover, this special issue also welcomes papers that focus on:  

    New applications of generative and agentic AI in large scale 3D reconstruction and novel view synthesis
    New applications on active acquisition and reconstruction
    Novel Datasets for challenging 3D reconstruction and rendering  

We encourage submissions that cover a broad range of visual reconstruction and view synthesis tasks. This special issue will provide a platform for researchers to share their latest findings and contribute to the advancement of photorealistic 3D reconstruction and novel view synthesis at-scale and in-the-wild. The contributions in this special issue could significantly benefit society by enabling more robust and reliable visual reconstruction algorithms, with industrial and commercial applications in AR/VR, robotics, and Spatial Intelligence.   

Timeline

    Final submission deadline: 15 January 2026
    First review decision: 15 April 2026
    Revised paper due: 15 May 2026
    Final review decision: 1 August 2026
    Final manuscript decision: September 2026