Image tokenizers map images to sequences of discrete tokens, and are a crucial component of autoregressive transformer-based image generation. The tokens are typically associated with spatial locations in the input image, arranged in raster scan order, which is not ideal for autoregressive modeling. In this paper, we propose to tokenize the image spectrum instead, obtained from a discrete wavelet transform (DWT), such that the sequence of tokens represents the image in a coarse-to-fine fashion. Our tokenizer brings several advantages: 1) it leverages that natural images are more compressible at high frequencies, 2) it can take and reconstruct images of different resolutions without retraining, 3) it improves the conditioning for next-token prediction -- instead of conditioning on a partial line-by-line reconstruction of the image, it takes a coarse reconstruction of the full image, 4) it enables partial decoding where the first few generated tokens can reconstruct a coarse version of the image, 5) it enables autoregressive models to be used for image upsampling. We evaluate the tokenizer reconstruction metrics as well as multiscale image generation, text-guided image upsampling and editing.

A current limitation of video generative video models is that they generate plausible looking frames, but poor motion — an issue that is not well captured by FVD and other popular methods for evaluating generated videos. Here we go beyond FVD by developing a metric which better measures plausible object interactions and motion. Our novel approach is based on auto-encoding point tracks and yields motion features that can be used to not only compare distributions of videos (as few as one generated and one ground truth, or as many as two datasets), but also for evaluating motion of single videos. We show that using point tracks instead of pixel reconstruction or action recognition features results in a metric which is markedly more sensitive to temporal distortions in synthetic data, and can predict human evaluations of temporal consistency and realism in generated videos obtained from open-source models better than a wide range of alternatives. We also show that by using a point track representation, we can spatiotemporally localize generative video inconsistencies, providing extra interpretability of generated video errors relative to prior work. An overview of the results and link to the code can be found on the project page: trajan-paper.github.io.

Classical light field rendering for novel view synthesis can accurately reproduce view-dependent effects such as reflection, refraction, and translucency, but requires a dense view sampling of the scene. Methods based on geometric reconstruction need only sparse views, but cannot accurately model non-Lambertian effects. We introduce a model that combines the strengths and mitigates the limitations of these two directions. By operating on a four-dimensional representation of the light field, our model learns to represent view-dependent effects accurately. By enforcing geometric constraints during training and inference, the scene geometry is implicitly learned from a sparse set of views. Concretely, we introduce a two-stage transformer-based model that first aggregates features along epipolar lines, then aggregates features along reference views to produce the color of a target ray. Additionally, we propose modifications that allow the model to generalize to scenes without any fine-tuning. Our model outperforms the state-of-the-art on multiple forward-facing and 360$^\circ$∘ datasets, with larger margins on scenes with severe view-dependent variations.

We propose a method to decompose a video into a background and a set of foreground layers, where the background captures stationary elements while the foreground layers capture moving objects along with their associated effects (e.g. shadows and reflections). Our approach is designed for unconstrained monocular videos, with an arbitrary camera and object motion. Prior work that tackles this problem assumes that the video can be mapped onto a fixed 2D canvas, severely limiting the possible space of camera motion. Instead, our method applies recent progress in monocular camera pose and depth estimation to create a full, RGBD video layer for the background, along with a video layer for each foreground object. To solve the underconstrained decomposition problem, we propose a new loss formulation based on multi-view consistency. We test our method on challenging videos with complex camera motion and show significant qualitative improvement over current approaches.

Classical light field rendering for novel view synthesis can accurately reproduce view-dependent effects such as reflection, refraction, and translucency, but requires a dense view sampling of the scene. Methods based on geometric reconstruction need only sparse views, but cannot accurately model non-Lambertian effects. We introduce a model that combines the strengths and mitigates the limitations of these two directions. By operating on a four-dimensional representation of the light field, our model learns to represent view-dependent effects accurately. By enforcing geometric constraints during training and inference, the scene geometry is implicitly learned from a sparse set of views. Concretely, we introduce a two-stage transformer-based model that first aggregates features along epipolar lines, then aggregates features along reference views to produce the color of a target ray. Our model outperforms the state-of-the-art on multiple forward-facing and 360° datasets, with larger margins on scenes with severe view-dependent variations. Code and results can be found at light-field-neural-rendering.github. io.

Traditional scene graph generation methods are trained using cross-entropy losses that treat objects and relationships as independent entities. Such a formulation, however, ignores the structure in the output space, in an inherently structured prediction problem. In this work, we introduce a novel energy-based learning framework for generating scene graphs. The proposed formulation allows for efficiently incorporating the structure of scene graphs in the output space. This additional constraint in the learning framework acts as an inductive bias and allows models to learn efficiently from a small number of labels. We use the proposed energy-based framework†to train existing stateof-the-art models and obtain a significant performance improvement, of up to 21% and 27%, on the Visual Genome [9] and GQA [5] benchmark datasets, respectively. Furthermore, we showcase the learning efficiency of the proposed framework by demonstrating superior performance in the zero- and few-shot settings where data is scarce.

Scene graph generation has emerged as an important problem in computer vision. While scene graphs provide a grounded representation of objects, their locations and relations in an image, they do so only at the granularity of proposal bounding boxes. In this work, we propose the first, to our knowledge, framework for pixel-level segmentation-grounded scene graph generation. Our framework is agnostic to the underlying scene graph generation method and address the lack of segmentation annotations in target scene graph datasets (e.g., Visual Genome [24]) through transfer and multi-task learning from, and with, an auxiliary dataset (e.g., MS COCO [29]). Specifically, each target object being detected is endowed with a segmentation mask, which is expressed as a lingual-similarity weighted linear combination over categories that have annotations present in an auxiliary dataset. These inferred masks, along with a Gaussian masking mechanism which grounds the relations at a pixel-level within the image, allow for improved relation prediction. The entire framework is end-to-end trainable and is learned in a multi-task manner. Code is available at github.com/ubc-vision/segmentation-sg.

Understanding images beyond salient actions involves reasoning about scene context, objects, and the roles they play in the captured event. Situation recognition has recently been introduced as the task of jointly reasoning about the verbs (actions) and a set of semantic-role and entity (noun) pairs in the form of action frames. Labeling an image with an action frame requires an assignment of values (nouns) to the roles based on the observed image content. Among the inherent challenges are the rich conditional structured dependencies between the output role assignments and the overall semantic sparsity. In this paper, we propose a novel mixture-kernel attention graph neural network (GNN) architecture designed to address these challenges. Our GNN enables dynamic graph structure during training and inference, through the use of a graph attention mechanism, and context-aware interactions between role pairs. We illustrate the efficacy of our model and design choices by conducting experiments on imSitu benchmark dataset, with accuracy improvements of up to 10% over the state-of-the-art.