Research map / Video / Video segmentation and tracking
On the choice of Perception Loss Function for Learned Video Compression
Sadaf Salehkalaibar, Truong Buu Phan, Jun Chen, Wei Yu, Ashish J Khisti
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use PLF-JD instead of PLF-FMD | MovingMNIST causal video compression · MovingMNIST, R1=12 bits, R2=12 | MSE distortion: 1.43±0.02 -> 2.26 ± 0.03 | Hurt |
| Use PLF-JD instead of PLF-FMD | MovingMNIST causal video compression · MovingMNIST, R1=12 bits, R2=∞ | MSE distortion: 0.0 -> 2.18 ± 0.02 | Hurt |
| Use PLF-JD instead of PLF-FMD | MovingMNIST causal video compression · MovingMNIST, R1=∞ bits, R2=1 | MSE distortion: 1.74 ± 0.02 -> 2.05 ± 0.03 | Hurt |
| Use PLF-JD instead of PLF-FMD | MovingMNIST causal video compression · MovingMNIST, R1=∞ bits, R2=3.17 | MSE distortion: 0.76 ± 0.01 -> 0.79 ± 0.01 | No clear effect |
| Use PLF-JD to preserve the joint frame distribution | Causal video decoder · MovingMNIST, low-rate GOP of 3 frames | temporal consistency | Improved |
| Use PLF-JD rather than PLF-FMD | Causal video decoder · MovingMNIST and UVG, low-rate initial frame | error correction across frames | Hurt |
| Train the encoder for MSE and choose the perception objective at the decoder | Video compression representations · MovingMNIST, rate-distortion-perception tradeoff | distortion-perception tradeoff | No clear effect |
| Convert an MMSE reconstruction to a reconstruction satisfying zero-PLF-FMD | Fixed video encoder · General sources | MSE distortion: at most a factor of 2 increase | Hurt |
| Use zero-PLF-JD for a general fixed encoder rather than MSE alone | Fixed video encoder · General sources | MSE distortion | Hurt |
| Adapt an MMSE representation using decoder-side linear transforms and possibly additive noise | Gauss-Markov video compression · Gauss-Markov source, any point in the iDP rate region | rate-distortion-perception tradeoff | Improved |
Closest papers by meaning
- Universal Rate-Distortion-Perception Representations for Lossy Compression (NeurIPS 2021 · similarity 0.90)
- Towards image compression with perfect realism at ultra-low bitrates (ICLR 2024 · similarity 0.86)
- Good, Cheap, and Fast: Overfitted Image Compression with Wasserstein Distortion (CVPR 2025 · similarity 0.86)
- Optimal Neural Compressors for the Rate-Distortion-Perception Tradeoff (NeurIPS 2025 · similarity 0.85)
- Controllable Distortion-Perception Tradeoff Through Latent Diffusion for Neural Image Compression (AAAI 2025 · similarity 0.85)
- Decouple Distortion from Perception: Region Adaptive Diffusion for Extreme-low Bitrate Perception Image Compression (CVPR 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
