12 papers
JLD: Perceptual Distance Through A Jacobian Lens
Shreshth Saini, Balu Adsumilli, Alan C. Bovik
Image compression, restoration, and generation all require a way to measure how different two images look to a person. Pixel error ignores how people see, while the most accurate p…
MEND: RL For Flow Models via Proximal Velocity Matching
Shreshth Saini, Neil Birkbeck, Yilin Wang +2
Reward post-training of flow models either reweights the model's own samples under a KL penalty or a frozen reference, often for thousands of updates, or backpropagates the reward…
LumaGuide: Distribution Shaping for Training-Free HDR Generation in Diffusion Models
Bowen Chen, Shreshth Saini, Balu Adsumilli +1
Pretrained diffusion models generate realistic images but are constrained by the statistical biases of their training data, limiting their ability to produce high dynamic range (HD…
CachedSearch: Training-Free Cached Exploration for Test-Time Search in Video Diffusion
Shreshth Saini, Neil Birkbeck, Yilin Wang +2
Test-time search lets small video diffusion models rival larger ones, but costs 2-10x more. All candidates are fully denoised, although most are discarded. Training-free caching ma…
GameScope: A Multi-Attribute, Multi-Codec Benchmark Dataset for Gaming Video Quality Assessment
Rajesh Sureddi, Shreshth Saini, Avinab Saha +1
The development of video game streaming has grown rapidly, with major platforms such as YouTube and Twitch using different codecs. To support quality assessment models that work co…
LumaFlux: Lifting 8-Bit Worlds to HDR Reality with Physically-Guided Diffusion Transformers
Shreshth Saini, Hakan Gedik, Neil Birkbeck +3
The rapid adoption of HDR-capable devices has created a pressing need to convert the 8-bit Standard Dynamic Range (SDR) content into perceptually and physically accurate 10-bit Hig…
Seeing Beyond 8bits: Subjective and Objective Quality Assessment of HDR-UGC Videos
Shreshth Saini, Bowen Chen, Neil Birkbeck +3
High Dynamic Range (HDR) user-generated (UGC) videos are rapidly proliferating across social platforms, yet most perceptual video quality assessment (VQA) systems remain tailored t…
TABES: Trajectory-Aware Backward-on-Entropy Steering for Masked Diffusion Models
Shreshth Saini, Avinab Saha, Balu Adsumilli +3
Masked Diffusion Models (MDMs) have emerged as a promising non-autoregressive paradigm for generative tasks, offering parallel decoding and bidirectional context utilization. Howev…
CHUG: Crowdsourced User-Generated HDR Video Quality Dataset
Shreshth Saini, Alan C. Bovik, Neil Birkbeck +2
High Dynamic Range (HDR) videos enhance visual experiences with superior brightness, contrast, and color depth. The surge of User-Generated Content (UGC) on platforms like YouTube…
Rectified-CFG++ for Flow Based Models
Shreshth Saini, Shashank Gupta, Alan C. Bovik
Classifier-free guidance (CFG) is the workhorse for steering large diffusion models toward text-conditioned targets, yet its native application to rectified flow (RF) based models…
Latent Guidance in Diffusion Models for Perceptual Evaluations
Shreshth Saini, Ru-Ling Liao, Yan Ye +1
Despite recent advancements in latent diffusion models that generate high-dimensional image data and perform various downstream tasks, there has been little exploration into percep…
HIDRO-VQA: High Dynamic Range Oracle for Video Quality Assessment
Shreshth Saini, Avinab Saha, Alan C. Bovik
We introduce HIDRO-VQA, a no-reference (NR) video quality assessment model designed to provide precise quality evaluations of High Dynamic Range (HDR) videos. HDR videos exhibit a…