Minsik Jeon

Yu Wu, Minsik Jeon, Jen-Hao Rick Chang et al.

We study positional encodings for multi-view transformers that process tokens from a set of posed input images, and seek a mechanism that encodes patches uniquely, allows SE(3)-invariant attention with multi-frequency similarity, and can be adaptive to the geometry of the underlying scene. We find that prior (absolute or relative) encoding schemes for multi-view attention do not meet the above desiderata, and present RayRoPE to address this gap. RayRoPE represents patch positions based on associated rays but leverages a predicted point along the ray instead of the direction for a geometry-aware encoding. To achieve SE(3) invariance, RayRoPE computes query-frame projective coordinates for computing multi-frequency similarity. Lastly, as the 'predicted' 3D point along a ray may not be precise, RayRoPE presents a mechanism to analytically compute the expected position encoding under uncertainty. We validate RayRoPE on the tasks of novel-view synthesis and stereo depth estimation and show that it consistently improves over alternate position encoding schemes (e.g. 15% relative improvement on LPIPS in CO3D). We also show that RayRoPE can seamlessly incorporate RGB-D input, resulting in even larger gains over alternatives that cannot positionally encode this information.

Zhongxiao Cong, Qitao Zhao, Minsik Jeon et al.

Current feed-forward 3D/4D reconstruction systems rely on dense geometry and pose supervision -- expensive to obtain at scale and particularly scarce for dynamic real-world scenes. We present Flow3r, a framework that augments visual geometry learning with dense 2D correspondences (`flow') as supervision, enabling scalable training from unlabeled monocular videos. Our key insight is that the flow prediction module should be factored: predicting flow between two images using geometry latents from one and pose latents from the other. This factorization directly guides the learning of both scene geometry and camera motion, and naturally extends to dynamic scenes. In controlled experiments, we show that factored flow prediction outperforms alternative designs and that performance scales consistently with unlabeled data. Integrating factored flow into existing visual geometry architectures and training with ${\sim}800$K unlabeled videos, Flow3r achieves state-of-the-art results across eight benchmarks spanning static and dynamic scenes, with its largest gains on in-the-wild dynamic videos where labeled data is most scarce.

Sunoh Lee, Minsik Jeon, Jihong Min et al.

Open World Object Detection(OWOD) addresses realistic scenarios where unseen object classes emerge, enabling detectors trained on known classes to detect unknown objects and incrementally incorporate the knowledge they provide. While existing OWOD methods primarily focus on detecting unknown objects, they often overlook the rich semantic relationships between detected objects, which are essential for scene understanding and applications in open-world environments (e.g., open-world tracking and novel class discovery). In this paper, we extend the OWOD framework to jointly detect unknown objects and learn semantically rich instance embeddings, enabling the detector to capture fine-grained semantic relationships between instances. To this end, we propose two modules that leverage the rich and generalizable knowledge of Vision Foundation Models(VFMs) and can be integrated into open-world object detectors. First, the Unknown Box Refine Module uses instance masks from the Segment Anything Model to accurately localize unknown objects. The Embedding Transfer Module then distills instance-wise semantic similarities from VFM features to the detector's embeddings via a relaxed contrastive loss, enabling the detector to learn a semantically meaningful and generalizable instance feature. Extensive experiments show that our method significantly improves both unknown object detection and instance embedding quality, while also enhancing performance in downstream tasks such as open-world tracking.

Minsik Jeon, Junwon Seo, Jihong Min

Despite the success of deep learning-based object detection methods in recent years, it is still challenging to make the object detector reliable in adverse weather conditions such as rain and snow. For the robust performance of object detectors, unsupervised domain adaptation has been utilized to adapt the detection network trained on clear weather images to adverse weather images. While previous methods do not explicitly address weather corruption during adaptation, the domain gap between clear and adverse weather can be decomposed into two factors with distinct characteristics: a style gap and a weather gap. In this paper, we present an unsupervised domain adaptation framework for object detection that can more effectively adapt to real-world environments with adverse weather conditions by addressing these two gaps separately. Our method resolves the style gap by concentrating on style-related information of high-level features using an attention module. Using self-supervised contrastive learning, our framework then reduces the weather gap and acquires instance features that are robust to weather corruption. Extensive experiments demonstrate that our method outperforms other methods for object detection in adverse weather conditions.

Qi Wu, Khiem Vuong, Minsik Jeon et al.

We tackle the problem of sparse novel view synthesis (NVS) using video diffusion models; given $K$ ($\approx 5$) multi-view images of a scene and their camera poses, we predict the view from a target camera pose. Many prior approaches leverage generative image priors encoded via diffusion models. However, models trained on single images lack multi-view knowledge. We instead argue that video models already contain implicit multi-view knowledge and so should be easier to adapt for NVS. Our key insight is to formulate sparse NVS as a low frame-rate video completion task. However, one challenge is that sparse NVS is defined over an unordered set of inputs, often too sparse to admit a meaningful order, so the models should be $\textit{invariant}$ to permutations of that input set. To this end, we present FrameCrafter, which adapts video models (naturally trained with coherent frame orderings) to permutation-invariant NVS through several architectural modifications, including per-frame latent encodings and removal of temporal positional embeddings. Our results suggest that video models can be easily trained to "forget" about time with minimal supervision, producing competitive performance on sparse-view NVS benchmarks. Project page: https://frame-crafter.github.io/